Genstat 16 Crackling

Auditory information is widely used throughout the animal kingdom in both terrestrial and aquatic environments. Some marine species are dependent on reefs for adult survival and reproduction, and are known to use reef noise to guide orientation towards suitable habitat. Many others that forage in food-rich inshore waters would, however, benefit from avoiding the high density of predators resident on reefs, but nothing is known about whether acoustic cues are used in this context. By analysing a sample of nearly 700,000 crustaceans, caught during experimental playbacks in light traps in the Great Barrier Reef lagoon, we demonstrate an auditory capability in a broad suite of previously neglected taxa, and provide the first evidence in any marine organisms that reef noise can act as a deterrent. In contrast to the larvae of species that require reef habitat for future success, which showed an attraction to broadcasted reef noise, taxa with a pelagic or nocturnally emergent lifestyle actively avoided it. Our results suggest that a far greater range of invertebrate taxa than previously thought can respond to acoustic cues, emphasising yet further the potential negative impact of globally increasing levels of underwater anthropogenic noise.

Introduction Across the animal kingdom, acoustic information is frequently used in orientation, habitat selection and predator avoidance. Marine coastal habitats, for example, are characterised by a high level of biological and abiotic noise, and coral reefs are particularly noisy due to high densities of resident shrimps, urchins and fishes,. Underwater, sound has two major components: in the acoustic nearfield (confined to an area within 1 or 2 wavelengths) particle velocity dominates, while in the acoustic farfield, the propagating pressure wave component dominates,. These acoustic components are detected by animals in two ways: sensory hair-like receptors are used to detect one-way particle displacement of water in the nearfield, whereas membranous receptors are used for the detection of farfield two-way particle oscillations. While these sensory mechanisms are well understood for fish and marine mammals, there is a relative paucity of information on whether aquatic invertebrates can also detect and utilise acoustic cues.

Many benthic marine organisms undergo an early developmental stage at sea and must settle to suitable habitat for juvenile and adult life. A number of studies have shown that settlement-stage larvae of a broad range of coral reef fishes can detect, and are attracted to, the noises of coral reefs –. There is also evidence that the larvae of some crabs and fishes in temperate waters use acoustic cues from urchin-dominated reefs to detect and locate settlement sites (see ). In addition to species that settle to reefs, the surrounding waters are home to a diverse community of free-swimming organisms (many of them crustaceans) that do not dwell in reef habitats; rather, their chances of survival are likely to be greatly enhanced by avoiding such areas of high potential predation risk,. Selection might therefore be expected to act on these species to evolve an ability to detect and avoid reef noise, but this possibility has never been explored.

Here we use experimental playbacks and light traps in the waters of the Great Barrier Reef lagoon to test the responses to coral reef noise of a broad suite of tropical crustaceans with a range of life-history strategies. We predict that larval stages of taxa that inhabit reefs as adults will, if they can detect the sound, be attracted to reef noise. In contrast, we predict that both pelagic taxa (those that remain in the water column throughout their lives) and nocturnally emergent taxa (those that ascend into the water column at night, but spend the day hidden in soft benthic sediment) will, if they are capable of detecting it, be deterred by reef noise. Ethics Statement All work was carried out under the guidelines of the Ethics Committees of the Australian Institute of Marine Science and Lizard Island Research Station, and with permission from the Great Barrier Reef Marine Park Authority, Australia. The study was conducted between November 2001 and January 2002 at Lizard Island Research Station (14°40′S 145°28′E), Great Barrier Reef, Australia.

An investigation by the prosecutor's office and Township police alleges the following sequence of events: At about 11:15 p.m., July 16, 2002, Michael Janicki dialed 911. §§ 46b-15(a), 46b-38a(2) (2012); D.C.Code § 16-1001(7)(C) (LexisNexis 2012); N.M. § 30-3-11(B) (LexisNexis 2012); N.D.

We sampled for 34 nights using a pair of light traps which consisted of an 8 W fluorescent light housed in a clear Perspex box with one 1×25 cm entry slit on each side. These traps are highly effective for sampling mobile, photopositive fishes and crustaceans. Traps were attached to permanent moorings, 180 m apart and >500 m from shore, in 10–15 m depth of water over sand at one of three locations (two in front of the Research Station and one at Coconut Beach, location determined by prevailing weather conditions; ). Each night, one light trap was randomly allocated a sound system while the other had a dummy rig attached to eliminate modification of the catch arising from additional floating objects. Our sound system consisted of a waterproof barrel containing a 12 V marine battery, 70 W amplifier and portable CD player, playing back reef noise through an underwater speaker (UW-30, frequency response 0.1 to 10 kHz, University Sound, Buchanan) fixed 1 m below the water surface and 1 m from the light trap. We used a 4-min recording of reef noise (), made using a calibrated Clevite CH17 hydrophone (flat response between 1.1–15 kHz, 5 dB drop-off below 1 kHz), a RANRL preamplifier (40 dB gain) and a Sony TCD-D7 digital tape deck.

The recording was from a mid-shelf reef on the Great Barrier Reef which is similar to the reefs surrounding Lizard Island, and captured the dusk chorus of biological noise recorded during the new moon phase, consisting of a chorus of pops made by nocturnal fishes together with a higher frequency (2.5– >20 kHz) but lower intensity background crackle produced by snapping shrimps as well as other feeding, movement and calling sounds. The recording was played back throughout the night on a continuous loop at a broadband (root mean square, rms) playback level set at 104 dB re 1 µPa, which ensured that the trap without playback did not receive additional noise above local ambient sound levels (measured at dusk at rms level of 93.8 dB re 1 µPa; ). Using p = ρ cv (where p = pressure in Pa, ρ = water density in kg m −3, c = speed of sound in m s −1, and v = particle velocity in m s −1 ), the particle velocity near to the speaker during playback would be 6.68×10 −8 m s −1. Acoustic representation of the reef recording and experimental conditions. Traps were deployed at dusk and retrieved at dawn. The catch was preserved in 70% ethanol and the crustaceans separated from the fish prior to categorisation and counting. To ensure that the numbers of captured crustaceans had not been modified by fish predation in the traps, we dissected 90 pelagic baitfish and 90 settlement-stage reef fish selected evenly from the two sound treatments and randomly from nine different nights.

We found no fish with freshly consumed crustaceans in their mouth, throat or stomach, so rule out the possibility that differential predation drives any differences in crustacean catches. The vast majority (99.3%) of the nearly 700,000 crustaceans caught were divided into 15 reliably distinguishable categories using a dissecting microscope; the remainder were not included in analyses. Any categories for which more than 50% of nights produced no catch in both traps (implying that there were no individuals of this category in the location on that occasion) were discarded prior to analysis. This criterion eliminated Euphausiacea, Palinura and Stenopodidae.

Remaining categories for which the mean nightly catch was less than 200 individuals were also discarded. This criterion eliminated Isopoda, Sergestidae and Stomatopoda. We therefore had nine categories for statistical analysis: two larval developmental stages of reef-settling Brachyura (zoea and megalops), two pelagic taxa (Copepoda and Hyperiidea), and five taxa that tend to be mostly nocturnally emergent (Caridea, Cumacea, Gammaridea, Mysidae and Ostracoda). Data were analysed using generalised linear mixed models (GLMMs) to allow the inclusion of random as well as fixed terms and thus control for repeated measures from the same trap locations and paired trapping on the same night. For each crustacean category, we used a separate GLMM with a Poisson error distribution and a log link function to examine how sound treatment (reef noise playback; ambient-noise, no-playback control) affected number of individuals caught in the trap.

Each GLMM was based on 68 catch totals from paired trapping on 34 nights at three different locations. Variance components were estimated using the Restricted Maximum Likelihood (REML) method, and random terms were retained unless the variance component was found to be zero (and hence their removal did not influence the analysis). The significance of fixed terms was determined using the Wald statistic, which approximates the χ 2 distribution. In each model, we included trap pair (i.e.

The two traps from the same night) nested in trap location as a random term. Statistical analyses were two-tailed and were conducted in Genstat (13th edition, Lawes Agricultural Trust, Rothampstead, Harpenden, UK). Results Of the approximately 691,000 individuals analysed statistically, 9.3% were developmental stage reef-settling Brachyura (megalops: 5.7%; zoea: 3.6%), 18.9% were pelagic taxa (Copepoda: 1.7%, Hyperiidea: 17.2%), and 71.8% were taxa that tend to be mostly nocturnally emergent (Caridea: 2.4%; Cumacea: 12.6%; Gammaridea: 9.5%; Mysidae: 42.1%; Ostracoda: 5.3%). There was no significant difference in the number of brachyuran megalops caught depending on sound treatment (GLMM: Wald statistic = 3.05, df = 1, P = 0.085), but brachyuran zoea were caught in significantly higher numbers in traps playing back reef noise compared to control traps (Wald statistic = 5.63, df = 1, P = 0.021; ). Catches of crustacean taxa in light traps with and without reef noise playback.

Both pelagic taxa were found in significantly greater numbers in the control traps compared to those with reef noise playback (Copepoda: Wald statistic = 17.22, df = 1, P. Discussion Our study demonstrates that a wide range of crustaceans with a variety of habits and life-history strategies are capable of detecting and responding to acoustic information. Previous evidence for such behaviour is restricted to the larval stages of a subset of taxa (mostly crabs) that recruit to reef habitat,. Furthermore, we provide the first experimental evidence in any marine organisms that taxa found in the proximity of reefs, but which do not settle to them, actively avoid reef noise.

These taxa can potentially benefit from such avoidance behaviour because reefs are home to a wide variety of mobile and site-attached predators that feed there both day and night,. Exploitation of the rich resources available in inshore waters must be balanced against this risk of predation, and mechanisms for optimising this trade-off should be selected over evolutionary time. Sound provides an excellent indicator of the direction and proximity of reefs and there is a clear survival benefit in utilising acoustic information to detect and avoid such hazardous locations.

As predicted, zoea, the pre-settlement larval stage of Brachyura, were attracted to reef noise. This is consistent with findings from temperate waters (e.g. ), but provides the first evidence for such a response in the tropics. The ability of zoea to use acoustic cues to help locate and remain within the proximity of suitable settlement habitat could be critical for recruitment success, despite any increased predation pressures. In contrast to zoea, megalops, the larval settlement stage of Brachyura, appeared not to be attracted to reef noise. One possible explanation for this is that any attraction of these late-stage larvae to reef noise (see ) was countered by a downward-swimming settlement response induced by the same noise (see ), causing some megalops to move away from those traps coupled with noise playback.

In addition to a diverse suite of biological noises, the soundscape in shallow water environments is influenced by local bathymetry, seabed characteristics and surface conditions. These factors combine to determine the distance over which reef noise propagates above ambient offshore levels. Since hearing in crustaceans is poorly understood, and may be in the farfield via specialised acoustic pressure detectors or limited to the nearfield through particle motion detection, a broad taxonomic investigation of hearing mechanisms and thresholds is needed to enable predictions of the likely distance of detection of reef habitats by crustaceans. Coral reef noise is heterogeneous in time and space, and these differences relate directly to habitat type, and the density of fishes. The reef sounds we played back were largely comprised of a background crackle generated by snapping shrimp and the pops, grunts and gurgles of nocturnal fishes (predominantly Holocentriae and Apogonidae).

More work, potentially using in situ choice chambers,, is needed to determine the level of selectivity of crustaceans to different sounds, and whether specific sounds (e.g., predatory fish vocalisations) or general broadband noise levels drive their directional behaviour. There is much recent concern that natural marine soundscapes are being modified or dominated in some places by anthropogenic noise arising from, for example, shipping and small boats, drilling and mining, seismic surveys and offshore construction. In modified acoustic environments, this can lead to masking of naturally important cues which, given our results, may mean that reef-settling crustaceans detect suitable adult habitat over smaller distances, and non-settling crustaceans are less able to detect and avoid potentially dangerous reef environments.

In addition, a recent study has demonstrated that, following several hours of exposure, reef fish larvae can become attracted to artificial sounds that would normally be avoided. If this was also the case for crustaceans, anthropogenic noise could lead to maladaptive behaviour by invertebrate taxa that underpin critical foodwebs and fisheries. Our study, demonstrating detection and ecologically relevant use of reef noise in a broad suite of tropical crustaceans, suggests that the use of sound for orientation is far more widespread than previously thought, and highlights the need for further research into the impact of anthropogenic noise throughout marine ecosystems.

Full text of ' 3 2044 105 172 738 V, 15 HARVARD UNIVERSITY LIBRARY OF THE GRAY HERBARIUM Digitized by the Internet Archi in 2015 https://arch i ve. O rg/detai Is/watso n i a 1 534bota WATSONIA r JournaTan^^Trocee^ of the Botanical Society of the British Isles Volume 15 Part 3 February 1985 Editors: J. Robson ISSN: 0043-1532 Botanical Society of the British Isles Patron: Her Majesty Queen Elizabeth the Queen Mother Applications for membership should be addressed to the Hon. General Secretary, c/o Department of Botany, British Museum (Natural History), Cromwell Road, London, SW7 5BD, from whom copies of the Society's Prospectus may be obtained. Officers for 1984-85 Elected at the Annual General Meeting, 19th May 1984 President, Mr J. Cannon Vice-Presidents, Mr R.

Wallace, Dr S. Walters, Professor D.

Webb Honorary General Secretary, Mrs M. Briggs Honorary Treasurer, Mr M. Walpole Honorary Editors, Dr J. Akeroyd, Dr S.

Gornall, Dr N. Robson, Dr B. Rushton (Watsonia); Mr D. Kent {B.S.B.I.

Abstracts)- Mr E. Wiggins {B.S.B.I. News) Honorary Meetings Secretary, Mrs J. Robertson (nee Martin) Back issues of Watsonia are handled by Messrs Wm Dawson & Sons Limited, Cannon House, Folkestone, Kent, to whom orders for all issues prior to Volume 15 part 1 should be sent. Recent issues (Vol.

15 part 1 onwards) are available from the Hon. Treasurer of the B.S.B.I., 68 Outwoods Road, Loughborough, Leicestershire. Watsonia, 15, 177-181 (1985) 177 Presidential Address, 1984 J. CANNON SEABALLS AND LAKEBALLS-AN OLD MEDITERRANEAN THEME WITH A NEW IRISH VARIATION Successive Presidents of the B.

S.B.I, have repeatedly laid stress on the fruitful cooperation between amateurs and professionals that has been such an important feature of the Society for many years past. Most notably in recent years, Mr R. David at the 1980 Annual General Meeting in Cambridge gave a most elegant, urbane and wide-ranging review under the title 'Gentlemen and Players'. In my address this morning, I shall not attempt to emulate his approach, but I shall try to illustrate an aspect of the relationship between amateurism and professionalism that has not attracted much attention.

In so doing, I pay my tribute to this twofold dependency which contributes so much to the on-going vitality of the Society. The high standard of expertise achieved by some amateurs is fully worthy of comparison with the activities of professionals and, as such, has not infrequently been a source of comment both in our own field and in related areas of Natural History. We hear much less about the 'amateur' activities of professional botanists. By these I do not, of course, imply the perjorative sense in which the word amateur is so often used today, but rather the true meaning of the word conveying involvement with an activity solely for the pleasure it engenders and without thought of financial or other advantage. Activities undertaken solely because they are worthwhile and enjoyable are always especially pleasing; and botanizing for fun can help the sometimes jaded professional to relive the simple pleasures of earlier stages in a botanical life not least through holiday activities which can make the proverbial busman's holiday look recklessly wide-ranging by comparison. As with the 'professional amateurs', we are fortunate that our Society fosters many opportunities for a genuinely amateur love of botany amongst our professional membership.

My concern with seaballs has been very much an 'amateur' interest, and may perhaps illustrate my general theme, not least through the homely nature of the only equipment involved in the experimental phase of the investigation. My curiosity was aroused more than 30 years ago when, as a recently appointed member of the British Museum staff, I had to deal with an enquiry from a lady who had been on holiday to the Mediterranean. She had found a strange fibrous ball on the beach, about 8 cm in diameter and so regular as to appear man-made.

More experienced colleagues at once recognized it as a Posidonia ball which, in those days, were quite often the subject of public enquiries. They do not seem to engage the attention of today's holiday-makers so frequently; perhaps the natural wonders of the Mediterranean now seem commonplace in this era of mass-produced and packaged travel. In due course, I visited the Mediterranean myself and was delighted to find large numbers of these strange objects cast up on the beach. Conventional wisdom refers to the balls as being formed by wave action from the fibrous remains of Posidonia leaves, which sounds an easy explanation until you actually wonder how it happens -a topic to which I shall return later. Posidonia oceanica is one of the very few truly marine flowering plants and is related to the Zostera species that are familiar to most British field botanists. In the Mediterranean, Posidonia forms wide 'meadows' in the shallow water adjacent to the gently sloping sandy beaches.

The only real way to gain an impression of these communities is through the use of a diving mask and snorkel, and there can be few more relaxing ways of botanizing than floating gently over a Posidonia bed and observing the plants and numerous small animals that shelter in its often quite dense growth. The growth habit of Posidonia is reminiscent of a mmiature rhizomatous Iris, and the fibres from which the balls are made are gradually released from the eroding leaves and leaf bases.

Elsie Parry, an American author of a semi-popular article on seaballs wrote: 'The Mediterranean Sea has long been the producer of fascinating wonders. The seaball is one of its 'Grade B' productions that has been appearing since ancient times. Greek and Roman writers mentioned 'bodies' that had been 'abandoned' by the sea. Galen and Aristotle wrote about 178 J. CANNON using the ashes of seaballs as a cure for scrofula, and in 1837 Germain de Saint-Pierre reported that hunters in Provence used the balls to wad their guns. Today, in districts where they are plentiful, seaballs are used commercially in the manufacture of paper and mattresses' (Parry 1956).

Although the balls sometimes occur in vast numbers, I am not aware of any current commercial exploitation and suspect that, in these days of extreme industrialization, the economic utilization of the balls may be a thing of the past. My concern with seaballs gradually led me to take a somewhat casual interest in the wider aspects of ball formation in the plant kingdom, and in this I have been greatly assisted by my colleagues with specialist knowledge of the groups concerned. Some of the strangest instances of ball formation are found in the lower cryptogamic groups, such as the green and red algae and the mosses. Unlike the Posidonia balls, which are composed of dead plant detritus, the algal and moss balls consist of living plants and while, like Posidonia.

They owe their origin to purely physical factors, their maintenance and growth is partly a result of the activities of the living plant. The best known of these are formed in both sea and freshwater by species of the green filamentous alga Cladophora, and in these and similar cases the balls are known as aegagropilous forms of the normal species. On occasions they can occur in great abundance and may even attract sensational headlines in the media, as when in 1950 an article appeared in the Illustrated London News entitled 'A beach ball mystery at Torbay'. On this occasion a 10 ft wide belt at least a mile long, containing many millions of balls, was formed on the beach (Newton 1950).

A similar occurrence was reported as an extensive deposit of 'sea manure' in 1903 at Lynn Beach in Massachusetts. Cladophoras have also attracted attention through their tendency to form balls in fresh water. These have received special notice in Japan, where a particular lake -Lake Akan, Hokkaido -has even been designated as a 'Special Natural Monument of Japan', and the balls have received the ultimate accolade -depiction on a postage stamp. The balls are formed by water movements from growing filaments and gradually become larger, though the annual incremental growth is often very slow, being as little as 5-10 mm in large balls which have been compared to footballs in size. In section, the balls often show indistinct concentric rings, which may be analogous to the annual growth rings of trees, although there does not s^em to be any very firm basis for this interpretation. The centre of old balls frequently decays to a greater or lesser extent and, since the filaments that form the walls of the Cladophora balls are living, they photosynthesize and respire like any other chlorophyll-containing plant tissue. As a result, their density may be altered by the release of gases from these metabolic activities, and consequently the balls rise or fall in the water in patterns that may be related to the diurnal physiological activity of the cells.

It may be this factor that has specially contributed to the popular interest of the Japanese and others in the lakeball phenomenon, and those in Lake Akan were apparently venerated by the local tribe in times past. Lakeballs were first recorded as long ago as 1588, when they were referred to by Ole Worm, alias Wormius, a Danish doctor of medicine, who called them 'pilla aquatica'. The first record in Britain appears to come from the letters of Thomas Knowlton, a distinguished gardener and plantsman who in the earlier part of the 18th century was corresponding with various men of science and learning. Some species of red seaweeds lay down calcareous deposits and can contribute to reef formation somewhat in the manner of corals.

Surprisingly, calcified algal protruberances can become detached and can then go on growing as calcareous algal balls. The continued growth of the whole ball is, of course, dependent on continued movement by waves and tidal currents, so that all aspects of the balls are more or less equally exposed to sunlight. In some places such as Brittany, Cornwall and parts of the western coast of Ireland, these balls are so numerous as to form beds or bands known as Maerl. The Maerl beds of Brittany have been exploited by dredging as a commercial source of lime for agricultural purposes (Blunden et al.

The whole topic of unattached seaweeds has been ably reviewed by Norton & Mathieson (1983). Ball formation by semi-aquatic mosses has also been quite frequently recorded, as in a recent paper by Deguchi & Inoue (1982), in which balls formed by mosses of the genera Blindia and Drepanocladus were noted from lakes in Tierra del Fuego and southern Chile.

As with the algae, the balls seem to have functioned as living organisms. Digressing briefly to consider similar structures and functions in terrestrial plants, we may recall that moss balls on land have been described from a wide range of habitats, from forest floors, through sandy and gravelly barren ground to the surface of glaciers.

In a recently described example from wetland forest near Amstelveen in the Province of North Holland, Wiegers (1984) discovered PRESIDENTIAL ADDRESS, 1984 179 that the motive power that kept balls, composed of the mosses Dicranum scoparium and Mnium hornum, in motion was the eager beak movements of foraging pheasants searching for food on the forest floor. Balls composed of Leucobryum glaucum are not infrequently seen in places where the moss can grow relatively undisturbed by human activities.

The occurrence of moss balls on ablating glacier ice is sufficiently well-known to have achieved a vernacular name - 'glacier mice', no doubt a reference to their round, furry appearance. In this case, the general instability of the habitat, coupled with the rigorous climate, tends to maintain mobility and, at the same time, promotes the colonization of newly opened habitats (Richardson 1981). Somewhat similar life forms are known in lichens (Weber 1977) where they are sometimes referred to as 'vagrants'. Examples are often found in the genera Aspilicia and Chrondropsis in high, open, windy areas in Asia and the manna which the Bible records as saving the starving Israelites in the desert has been ascribed to a similar source.

Under moist conditions the lichens may open out, curling into a ball again with the onset of arid conditions. This may be a means of dispersal and as such is reminiscent of the tumbleweeds, a life form that includes plants as various as a Selaginella and a heterogeneous range of flowering plants of open habitats. These species grow under desert-like conditions, and during droughts roll into balls, which may become detached and roll around in the wind. In the course of this movement, seeds may be dispersed and eventually, with the return of moist conditions, the plants open out and may become reattached in some favourable situation for further growth. Such plants are sometimes seen in tourist and curio shops under names such as Rose of Jericho or Resurrection Plant. This is perhaps enough general background and I must now return to my major theme of Posidonia balls.

When explaining to enquirers how the balls are formed, I have frequently found myself sharing their incredulity that the simple physical forces of wave and current, acting on a supply of plant fibres, could result in such remarkably uniform objects, so as to cause their finders to believe them to be man-made. As a starting point for understanding the processes involved, we should remember that animal and plant fibres which to us seem smooth, are actually quite rough at the microscopical level, with minute scales.

Without these, wool from animals and flax and cotton from plants could not be arranged in an orderly manner during the process of spinning, and then remain twisted and adhering together to form usable threads. Likewise, and perhaps here the parallel is closer to Posidonia, wet wool can be manually compacted in such a way as to produce a stable fabric -felt.

But at the spinner's wheel or spindle we are still a long way from explaining what happens at the bottom of the sea. The nagging doubt as to how the balls were formed remained until a few years ago, when I noticed that our domestic washing machine (Hotpoint Automatic De Luxe 1972 -a 'top loader') was producing small objects, quite reminiscent of Posidonia balls, from the miscellaneous textile fibres that became detached from clothing in the wash. Most, if not all, washing machines have a lint-trap to collect these fibres and the machine in question had a trap in the form of an oscillating plastic sieve, 21.5 cm in diameter and 3 cm deep. The sieve, which has numerous fine drainage holes, is situated on the top of the main agitator, and oscillates in the horizontal plane through c.

90° at a rate of 80 cycles per minute. Water from the main washing drum is circulated through a spout above the sieve while the main washing programme is in operation, so that fibres which might otherwise clog the machine are filtered from the water. Here at last was the moment of breakthrough! If the washing machine could make incipient balls from clothes fibres, could it also make an artificial Posidonia ball from Posidonia fibres? With some trepidation we decided to carry out some experiments, fearful lest the bowels of the machine were at risk from this unaccustomed diet. Fortunately, the machine would operate during the washing phase with the lid open, so it was possible to observe what took place.

I am glad to be able to record that it survived the ordeal to give further years of faithful service. A ball from Giens in southern France was carefully broken down by hand to its constituent fibres.

These were placed in the main drum and, when operations were started, gradually made their appearance in the filter tray. Soon some of the fibres started clumping together in loose, soggy masses of about 1-2x0.5 cm. These initials were very fragile so that, if two or more came into contact, disintegration often occurred, although rarely, amalgamation took place.

The rate of availability of fibres is an important factor. This is not a simple process of accretion, as when a snowball is rolled along the ground, and our observations showed that if fibres are too abundant, ball formation tends not to occur, as the numerous initials interact and mutually disrupt at early stages. When a ball has grown to the initial stage, it is still very delicate and loosely constructed. It is now essential for the second phase of ball formation to come into play. The ball is now gradually 180 J. CANNON consolidated towards the firm consistency with which we are familiar.

This results from backwards and forwards rolling and, for this to occur, the nature of the substrate is important so that, when the water changes direction, the friction between the ball and the sandy bottom results in a rolling rather than a skidding motion. In our experimental model, the holes on the bottom of the sieve acted in a manner analogous to the sand grains on the sea bed. Under these experimental conditions, three washing cycles of ten minutes produced a reasonably compacted ball of 5.0x3.5 cm. While the experiment demonstrated some of the factors involved in ball formation, no real light is shed on the time involved under actual seabed conditions (Cannon 1979). So much then for the classic Mediterranean theme; what about the new Irish variation? Last year when Mr A.

Chater, Dr S. Walters and Prof.

Webb were walking on the Inch sand spit which projects southwards across Dingle Bay in Kerry, some 35 km from its seaward end, they encountered a fibrous structure remarkably similar to a Posidonia ball. So far, efforts to determine its origin and composition have not been very successful. I am indebted to Dr Paula Rudall of the Jodrell Laboratory of the Royal Botanic Gardens, Kew, who kindly made an anatomical examination of the balTs fibres. She reports that they are the central vascular portions of roots, but with the cortex and epidermis lacking. They are most likely from a Monocotyledon, as they are polyarch with no secondary growth, and have a multilayered pericycle which occurs in many Monocotyledons such as grasses. But as primary root structures do not vary very much, and published data and reference materials are not very extensive, a more detailed identification was not possible.

Where did it come from? Correspondence with Dr E. Nelson of the National Botanic Gardens, Glasnevin, Dublin, who has made extensive studies of drift seeds brought by the Gulf Stream, failed to produce evidence of any comparable occurrences. In his opinion, a fibrous structure of this kind would be unlikely to remain afloat for the fifteen or more months needed to cross the Atlantic from the Caribbean, even supposing it had a dry terrestrial origin. So we are left with another, albeit minor, enigma in the flora of our islands.

Members are invited to keep their eyes open while walking on Irish and British beaches. Was the Chater/Walters/Webb ball a unique occurrence?

Clearly much remains to be discovered about plant fibre balls. For instance, in the Kew Museum there are balls formed 'from larch leaves felted together by the agitation of water, taken from the bottom of lakes of Shropshire and in Bagshot Park'. In the British Isles we cannot hope to rival the prodigious ball productivity of the Mediterranean, but there may yet be matters of interest to discover in this somewhat unusual field of Natural History. REFERENCES Bi.UNDLN, G., BiNNS, W. Commercial collection and utilisation of Maerl.

Bot., 29: 140-145. An experimental investigation of Posidonia balls. Bot., 6: 407-410. Lake moss-balls found in Tierra del Fuego and Brunswick Peninsula, southern South America. Mas., Tokyo, Ser. B (Botany), 8: 145-150.

Lake ball ^Marimo' in Lake Akan. PhycoL, 28: 168-169. A beach ball mystery at Torbay. News, 216: 98. & Mathieson, A. The biology of unattached seaweeds, in Round, F.

& Chapman, D. Progress in Phycological Research, 2: 333-386. Beachcomber's treasure-a sea ball. Hist., N.Y., 65: 496-497. Richardson, D. The biology of mosses.

Environmental modification and lichen taxonomy, in Seaward, M. Lichen ecology, pp. Observations on the origin of 'moss balls' in a Dutch wetland forest.

58: 449- 454. NOTE ADDED IN PROOF Since this address was given, my wife and I have had the opportunity of visiting the Inch sand spit in Dingle Bay, where Chater, Walters and Webb found the mystery seaball.

We were able to find a considerable number of the balls and to reach the conclusion that their major constituent was fibrous material from Ammophila arenaria roots, as had been suggested might be the case by Dr T. PRESIDENTIAL ADDRESS, 1984 181 G. Curtis (after the address was presented for a second time at the B. Irish Region Annual General Meeting on Saturday 8th September 1984). The question still remains, however, as to whether this locality is the only site in the British Isles where seaballs are produced from this apparently abundant raw material. J Watsonia, 15, 183-209 (1985) 183 Taxonomy of Atriplex species indigenous to the British Isles p.

TASCHEREAU* Department of Botany, University of Manchester ABSTRACT A biosystematic study of the genus Atriplex (Chenopodiaceae) based on field, culture, experimental hybridization, herbarium and cytological work delineates the taxa of this genus indigenous to the British Isles. Detailed morphological descriptions are given and distribution maps and illustrations provided for /i. Prostrata Boucher ex DC, A.

Glabriuscula Edmondston, A. Longipes Drejer, A.

Praecox Hiilphers, A. Littoralis L., A. Patula L., and A. The habitats and reproductive biology of the species are discussed and the chromosome numbers reported. Hybrid derivatives between A. Longipes and A. Glabriuscula and between A.

Longipes and A. Prostrata are common in many areas of the coast. The following hybrids are also reported: A. Glabriuscula x A. Glabriuscula x A.

Prostrata, A. Littoralis x A.

Prostrata, and A. Littoralis x A.

INTRODUCTION The most recent treatment of Atriplex occurring in the British Isles is that of Aellen (1964) in Flora Europaea. According to Aellen the following species are native: A. Laciniata L., A. Patula L., A. Littoralis L., A. Prostrata Boucher ex DC.

Glabriuscula Edmonston and A. Longipes Drejer.

Atriplex laciniata is placed in section Sclerocalymma Aschers., and the remaining species in section Teutliopsis Dum. Within this section A. Prostrata, A. Glabriuscula and A. Longipes form a recognizable unit, the A. Prostrata group or Hastata complex. The taxonomic problems in British Atriplex species have been concerned with the members of section Teutliopsis and in particular with the /I.

Prostrata group. In section Teutliopsis, the number of species recognized in British floristic works has varied from one in Bentham & Hooker (1896) to nine in Babington (1841). Druce (1928) recognized 21 native taxa of which six were treated by him as species and one was considered a hybrid. The considerable variation within the A. Prostrata group is reflected in the taxonomic treatment it has received from British authors.

Babington wrote that G. Bentham believed A. Glabriuscula to be indistinguishable from A. Prostrata (A.

Babington 1897). This difference of opinion resulted in different treatments of these plants in two important British floras -that of Babington (1843) and of Bentham & Hooker (1896).

Hulme (1957) did not distinguish between the coastal forms of the A. Prostrata group, referring to them as 'hastata-glabriuscula' Tutin (1962), while treating A.

Glabriuscula as a species distinct from A. Prostrata, added the remark that A. Glabriuscula is 'probably best regarded as a subspecies of it'.

Until recently only A. Prostrata and A. Glabriuscula were recognized as occurring in Britain. Hulme (Aellen 1964) mentioned ^4. Longipes, which was said to be 'widespread in the British Isles', but other authorities (Gustafsson 1972) doubted its presence here and no material was available for study.

Jones (1975) observed plants resembling A. Longipes in a few localities in Britain and she suggested that this taxon was present here. In 1977, 1 reported the presence oi A. Praecox Hiilphers in the British Isles and confirmed the presence of A. Longipes (Taschereau 1977). Taxonomic ranking within the A. Prostrata group, previously concerned only with A.

Glabriuscula, must now consider these two additional members of the group. *Present address: Institute for Resource and Environmental Studies, Dalhousie University, Halifax, Nova Scotia, Canada, B3H 3E2 184 P. TASCHEREAU Although many authors have written about Atriplex, relatively few have contributed much to our understanding. Moss & Wilmott (1914) provided a well-illustrated and helpful monographic account of the Atriplex species occurring in Britain. Aellen (1960) dealt with the taxonomy, distribution, ecology and economic uses of European Atriplex. And in 1964 he provided a revised account of their taxonomy and distribution. Biosystematic studies have been extremely important in elucidating the taxonomy of species in section Teutliopsis: Turesson (1922a, 1922b, 1925), in a series of pioneer studies, demonstrated the value of extensive cultivation experiments and experimental hybridization in understanding the complexities of this group.

Hulme (1957, 1958) produced the first controlled experimental hybrids in Atriplex thereby demonstrating the feasibility of this approach to the study of the genus. Her work was of key importance in understanding the taxonomy of section Teutliopsis in North America as well as in Britain. Hulme's experimental findings supported the taxonomic separation of the tetraploid A. Patula from the diploids A. Littoralis and A.

Prostrata and the maintenance of these taxa at the level of species. Bentham & Hooker (1896) had united them as intergrading variants o{ A. Although by 1957 most British and European authors were treating all three taxa as separate species, this was not the case in North America. Since Gray (1868), North American taxonomists had regarded the members of section Teutliopsis including A.

Prostrata and A. Littoralis to be intergrading varieties oiA. And they continued to do so until convinced otherwise by Hulme's data and subsequent studies (Taschereau 1972).

(The confusion resulting from this traditional taxonomic treatment is still evident in ecological papers from North America in which A. Patula is referred to as a halophyte and a component of salt marsh communities. In North America, as in Britain and elsewhere, A. Patula is a ruderal and colonizer of disturbed soil, relatively salt tolerant but not a halophyte and never a component of salt marsh communities.) Minor contributions to the experimental taxonomy of section Teutliopsis were made by van der Meijden (1970) in the Netherlands, and by Jones (1975a, 1975b) in Britain. Neither author did extensive cultivation experiments nor attempted to make hybrids.

The most extensive and important biosystematic studies in Atriplex are those by Gustafsson (1972, 1973a, 1973b, 1974, 1976) of the A. Prostrata group in Scandinavia. They provide a foundation for understanding the group as it occurs in the British Isles. Without Gustafsson's many experimental hybrid specimens for reference, accurate identification of the hybrid derivatives that comprise so much of the British coastal Atriplex flora would have been impossible. MATERIALS AND METHODS The present work is based on a four-year study of the genus in the field, laboratory and botanic garden. FIELD STUDIES Great Britain Areas within each of the major plant regions of Britain (Heath & Scott 1974) were examined at least once between 1974 and 1978. 75 sites covering 27 vice-counties were visited.

The areas included 61 coastal and estuarine habitat sites, three inland salt marshes and eleven inland sites on disturbed ground. The location of these sites is given in Fig. Samples were taken from 255 populations and collections made comprising approximately 1,270 pressed herbarium specimens.

All but three of my locality records cited in this paper are supported by voucher specimens filed in MANCH. In 1977, a comprehensive survey of the Atriplex taxa on the coasts of the British Isles was undertaken through the Botanical Society of the British Isles in cooperation with the Biological Records Centre, Monks Wood.

The 251 herbarium specimens collected by the participants and 80 data cards were sent to me for identification and checking. Data were obtained from 36 vice-counties representing all coastal regions except south-western England and the coasts of Ireland. The areas from which specimens were received are shown in Fig. Sweden Three coastal localities in the province of Skane in southern Sweden were examined with M.

Gustafsson in 1975: Bunkeflo and Skanor Harbour south of Malmo, and Torekov near Angelholm TAXONOMY OF ATRIPLEX IN THE BRITISH ISLES 185 on the south-western coast. The Torekov site was studied by Turesson (1925) in his work on the genus Atriplex, and all the localities visited provided material for the Atriplex investigations by Gustafsson (1976). In 1976, three localities were examined on the coasts of the Baltic Sea: Kalmar on the mainland, Stora Ror and Ottenby on the island of Oland. Ottenby is the restricted type locality of Atriplex hastata L. (Steam 1973).

In total, six Swedish sites were examined and three populations were sampled, the taxa comprising five species and four hybrids. TASCHEREAU CULTIVATION Large numbers of plants were cultivated to study regional variation and phenotypic plasticity. Plants were also cultivated to confirm identification, to facilitate observations on reproductive biology, to make artificial hybrids and to observe segregation in natural and artificial hybrid progeny. From 1974 to 1978, approximately 2,650 plants were cultivated either in the greenhouse or in outdoor plots in the botanic garden. Of these, about 1,280 were grown to maturity. These included 27 taxa comprising representatives of all the four European sections of the genus. For reference and study, 380 specimens of pressed plants consisting of stems, leaves, bracteoles and fruit were prepared from the cultivated material, as well as 75 specimens consisting only of bracteoles and fruit.

Seeds were sown in trays of John Innes Seed Compost in the autumn and placed in an unheated greenhouse over winter. In this way the dormancy encountered in several species was overcome. The seedlings were later transferred to individual pots containing John Innes Potting Compost. Dormancy in almost all seeds could be overcome by placing the moistened seeds in a controlled cycling environment (Ignaciuk & Lee 1980) where they were exposed to 9 hours dark at 10°C and 15 hours light at 30°C. Germination then occurred within two weeks. In a few cases, however, seeds in the controlled cycling environment germinated only after the seed coat was also removed.

CYTOLOGY The chromosomes are small (2-3 im), metacentric to submetacentric, and morphologically similar. The basic number in the genus is x=9.

Endomitosis, the formation of cells containing multiples of the normal somatic number, is a phenomenon encountered in Atriplex root-tips, and one that complicates chromosome counting. Diploid, tetraploid and octoploid cells are commonly present in the same root-tip preparation. Polyploid cells increase in number along the root-tip. They can be largely avoided in cytological preparations by utilizing only a minute section of the root behind the root-cap. Root-tips obtained from vigorous young plants cultivated in pots in the greenhouse were pre- treated overnight in 0.2 mM solution of 8-hydroxyquinoline at 5°C then fixed in absolute alcohol- glacial acetic acid (3:1) for 24 hours.

The material was then transferred to 70% alcohol and stored in the deep freeze until use. The tips were hydrolysed in IN HCl for 9 minutes at 60°C then stained in Feulgen for 2 hours.

The excised tip was tapped and squashed in lacto-propionic orcein. Meiotic preparations were made by fixing very young buds in a mixture of absolute alcohol, chloroform and glacial acetic acid (6:3:1) for 24 hours in the deep freeze.

The fixed buds were transferred to 70% alcohol before use then dissected out in 45% lacto-propionic acid and squashed in lacto-propionic orcein. HERBARIUM Descriptions and other taxonomic data presented in this study are based almost entirely on specimens I collected between 1974—1978 or collections made by others during that period and sent to me. Herbarium studies were used primarily to supplement and confirm these data and to estabHsh the extent of variation within taxa from widely separate geographic regions.

My distribution records of A. Prostrata, A. Littoralis and A. Laciniata are supplemented by data provided by the Biological Records Centre. Additional distribution records of A. Glabriuscula and A. Praecox are based only on herbarium specimens I identified.

Material from the following herbaria was studied: ABD, ANK, BM, C, CGE, DBN, E, K, LD, LIV, LIVU, MANCH, NMW, OXF, SLBI, TCD (abbreviations according to Kent & Allen (1984) and Holmgren et al. Hulme at LIV was also examined. Specimens are not cited in this paper. I have, however, annotated the entire holdings of the following major British and Irish herbaria: ABD, CGE, DBN, E, LIV, LIVU, TCD.

Also, approximately half of the large holdings of NMW have been annotated by me. DIAGNOSTIC CHARACTERS The diagnostic characters used in Atriplex identification differ from those used in other chenopod genera. The inflorescence provides few characters, and the flowers virtually none. Even seed surface TAXONOMY 0¥ ATRIP LEX IN THE BRITISH ISLES 187 sculpturing, a character found so useful for separating closely related species of Chenopodium (Cole 1961), has so far not proved helpful in Atriplex. Within a species, the leaf outline can vary from one biotype to another in the same habitat and from one node to the next on the same plant. Within a species, colour, vestiture and habit can vary due to genetic difference or because of environmental factors.

Leaf colour outline and vestiture can also change with the maturity of the plant. Younger leaves or those remaining on the plant at maturity can differ greatly from mature leaves or the earlier-formed leaves that may have dropped off the plant by maturity. Most characters are to some extent variable. In some cases, the phenotypic plasticity is such that particular environmental factors can cause plants of one species to resemble the phenotype of another, genetically different species.

The most important diagnostic characters are those of the lower principal leaves and of the fruiting unit. The latter consists of the seed with surrounding pericarp and the pair of bracteoles within which they are enclosed.

Separation of species in section Teutliopsis depends essentially on the existence of a consistent correlation between the characters of the lower leaves and those of the fruiting unit. Some species have a restricted range in Great Britain and some occupy relatively specific habitats. Information about the habitat and locality is particularly useful in this group. Special terms used in the key and descriptions are illustrated in Fig.

2 and explained along with the major diagnostic characters below. HABIT Atriplex may be erect with branches ascending or outspreading, or prostrate (decumbent or procumbent).

My experiments and those of Turesson (1919) indicate that environment, particularly light, nutrient and moisture, has a considerable effect on the habit of some species. In several species two variants exist: a hereditary prostrate kind and a modificatory prostrate kind.

In the latter, intense light induces a plagiotropic response in a normally erect plant; and in the former, shading causes the branches to turn upwards (Turesson 1919). Though variable in some species, the habit is often distinctive. Littoralis in Britain is consistently erect.

Glabriuscula is characteristically procumbent on exposed beaches, becoming decumbent to weakly erect when crowded at the landward margin of the beach, but also possessing a less common erect variant. Patula, commonly erect though often spindly and falling over, has hereditary and modificatory prostrate variants. Prostrata has erect and prostrate kinds.

LEAVES Lower Principal. The lower principal leaves are the earlier-formed leaves on the central axis in the middle to lower part of the plant, at approximately nodes 4 to 8 up from the base, between flowering and the development of mature fruit. They often differ considerably in size and form from the later- developing upper leaves and they frequently drop off before the bracteoles and seed are fully mature.

The lower leaves on.4. Longipes, for example, are elongate-triangular with basal lobes while the upper leaves that develop on older specimens are lanceolate to linear and frequently entire. By the time the plants reach maturity, the only leaves that remain on them may be the lanceolate to hnear ones. The leaf outline, although extremely variable within some species, is an essential diagnostic character. The outline of the lower principal leaves is less variable and it is these that are taxonomically significant. The leaf base angle of the lower principal leaves, cautiously employed, can provide a useful secondary character. It is too variable to be used alone to separate A.

Glabriuscula from A. Prostrata but other species have a characteristic base. Patula and A. Praecox, for example, the leaf base is cuneate and in ^4. Longipes it is cuneate to obtuse. The leaf base often serves as a useful character for detecting hybrids between species with truncate leaves and those with cuneate leaves.

Between the upper and lower leaves on the same plant, the leaf base angle is often very different, and in some individuals it varies greatly even from one node to the next. Leaf succulence has little diagnostic value, although some individuals oi A. Glabriuscula may have extremely succulent leaves. Specimens of the hybrid littoralis x A.

Prostrata ure usually extremely succulent, perhaps as a result of positive heterosis. Betacyanins (Dreiding 1961) are responsible for the red colour in Atriplex. As a taxonomic 188 P. TASCHEREAU 4 5 6 7 8 9 10 11 12 Figure 2.

Diagnostic characters used in Atriplex. Kranztypus leaf venation (section Sclerocalymmd) 2. Normal dicotyledonous leaf venation (section Teiitliopsis) 3. Forward-curving basal lobes; 4. Lingulate apex; 5. Bracteole margins united up to the middle (arrow); 6.

Bracteole margins united only at the base (arrow); 7. Stalked bracteole; 8. Bracteoles thin ot evenly- thickened; 9. Bracteoles spongy-thick from the middle to the base; 10. Seed radicle strongly up-pointing; 11. Seed radicle obliquely up-pointing; 12.

Seed radicle out-pointing. Character colour is of secondary importance. The red colour in some individuals may be the result of a genetic difference, but the development of red may depend largely on environmental factors.

Praecox in Britain is characteristically red but in A. Glabriuscula, reddish and entirely green plants commonly occur together in the same population. Littoralis may be green or reddish and in A.

Patula, a species which is usually entirely green, reddish strains occur. The leaves of A.

Longipes frequently tuin bright yellow at maturity but so do those of some of its hybrids with A. Young Atriplex leaves are covered with stalked, oblong, fluid-filled vesicular hairs. These, as the leaf matures, dry and form a scaly or mealy surface. As a taxonomic character, the density of the scales and their distribution on the leaf surface has limited usefulness.

For example, the abundant mealiness on the leaves of some coastal species gives them a distinctive whitish appearance. Patula, by contrast, has only a sparse, hardly discernible covering of fine mealy particles on the younger leaves. TAXONOMY OY ATRIP LEX IN THE BRITISH ISLES 189 Venation. Leaf venation in section rewr//op5/s consists of the normal dicotyledonous type. Immature and densely lepidote specimens in section Teutliopsis can be immediately distinguished from A. Laciniata, the only native species not in section Teutliopsis, by the leaf venation.

Laciniata possesses the highly distinctive Kranztypus venation (Fig. This becomes readily visible with a hand lens when the leaf is scraped with a knife-blade.

BRACTEOLES The breacteoles provide some of the most useful diagnostic characters. Within many species, however, there can be considerable variation. The variation may be due to genetic differences between biotypes of the same species or differences in the environmental factors acting on the same biotypes. Once the limits of variation are known within a species, the bracteole characters can be usefully employed. The following characters should be observed: Outline. In most species the outline is either triangular, rhombic, ovate or some combination of these shapes. In most species, the apex is usually either acute, acuminate, lingulate or produced to a thin foliose tip.

In most species, the base is truncate, obtuse, or cuneate. The base, especially of bracteoles occurring in the axils of leaves or branches, should be examined for the presence of a stalk, a variable but important character whose presence is usually essential for the identification of hybrids involving A. How far the bracteole margins are united up from the base is a major diagnostic character. The presence or absence of lateral angles is important.

The degree to which the lateral angles are developed, whether the development is unilateral or bilateral, and whether the lateral angles are rounded or pointed are important characters. The degree of toothing and the position of the teeth are also significant. Dorsal Surface. The dorsal surface is described as smooth, muricate or tuberculate, strongly reticulate-veined, 1-veined or obscurely-veined. Both smooth and tuberculate individuals may occur within one species, and the degree to which the veins are prominent varies within individuals of the same taxon, but once these limitations are known, the characters can be usefully employed. Inner Surface. A layer of distinctive, brown or silvery-brown, spongy tissue of large air-filled parenchyma lines the inner surface of the bracteoles of certain species.

Within a single taxon, this tissue may be undeveloped, developed only at the base, or developed over the entire surface. Patula, for example, it is never present, but it typically occurs in members of the A. Prostrata group.

The bracteoles are described as membranous, herbaceous or spongy. Membranous bracteoles, such as often occur in A. Praecox are so thin that the fruit, not merely its outline, can be seen within them.

The bracteoles of A. Glabriuscula, for example, are spongy, particularly in the basal portions. This spongy character is especially useful for distinguishing hybrids between A. Glabriuscula and species that have membranous or herbaceous bracteoles. SEEDS The types of seeds present, their size, shape and position within the pericarp, provide important diagnostic criteria.

On the same plant, two distinct seed types may occur: a brown type and a black type. The brown seeds are reddish to dark brown, usually larger than the black, flattened and disc-shaped, have a radicle that is distinctly produced, have a dull or pebbled-glossy surface with the radicle region often strongly striate-pebbled, and possess a softer outer coat. The black seeds in most species are generally smaller than the brown, are biconvex, have a radicle that is scarcely produced, have a smooth-glossy surface and possess a harder outer coat. In some individuals, these morphological distinctions in seed type are relatively constant, but most taxa have many individuals in which numerous morphologically intermediate seed types occur.

The relative abundance of the two seed types within the individual has slight taxonomic significance because the ratio may differ within strains of the same species. This is particularly true oi A. Patula and A. In these species it is common to find that one or the other seed type predominates within a particular strain of plants. The distinctions in radicle position and direction discussed below, as well as differences in seed shape, are much more apparent in the larger, brown and intermediate types than in the smaller black types. TASCHEREAU Radicle Position and Direction. The position of the seed radicle and its direction are important taxonomic characters.

Surrounding the seed is a loosely or firmly attached membranous pericarp with the vestiges of the style situated at the top. The outline of the radicle becomes visible on the seed margin when the lower part of the pericarp is teased away. The radicle may emerge from the middle portion of the seed margin ('median') or from the base of the seed ('basal'). The radicle apex may be directed upwards towards the style vestiges ('up-pointing'), or outwards at a right angle to the style axis ('out-pointing'), or at an angle between these directions ('obliquely up-pointing'). KEY TO THE SPECIES 1.

Leaf venation appearing (at xl2) as a conspicuous, dark green, reticulate pattern (Kranztypus) when the leaf surface is scraped with a sharp blade; bracteoles cartilaginous and hardened in the lower half; seed transversely elliptical, light brown 7. Leaf venation appearing (at x 12) as the normal dicotyledonous type and not showing a dark green reticulate pattern when the leaf surface is scraped with a sharp blade; bracteoles herbaceous, membranous or spongy-inflated, not hardened in the lower half; seed ovate to orbicular, dark brown or black 2 2. Lower leaves linear or lanceolate 3 2.

Lower leaves triangular or rhombic-ovate 4 3. The lower leaves linear without basal lobes; bracteoles ovate, thick, margins united only at the base, apices acute or Ungulate and frequently reflexed at maturity. Coastal halophyte 5.

Littoralis 3. The lower leaves lanceolate with forward-curving basal lobes; bracteoles rhombic, thin, margins united almost to the middle, apices acute to acuminate and never reflexed at maturity. Non-halophytic inland and coastal weed of disturbed ground d.A. Lower leaves rhombic-ovate; mature plants small (mostly 8-10 cm high). Restricted to lower littoral zone of coastal beaches A.

Lower leaves triangular; mature plants larger (mostly more than 20 cm high). Occurring throughout the littoral zone of coastal beaches and inland 5 5. Bracteoles rhombic, margins united up to the middle, spongy-thick from the middle to the base; seed radicle strongly up-pointing. Coastal halophyte 2.

Glabriuscula 5. Bracteoles ovate or triangular, margins united only at the base, thin or evenly thickened; seed radicle out-pointing or obliquely up-pointing. Coastal and inland 6 6.

Lower leaves deltoid-triangular, length less than twice the width, base truncate to subcordate (base angle more than 160°); axillary bracteoles infrequent, morphologically similar to terminal ones, thin or evenly thickened, none foliaceous; margins with pointed or rounded, weakly developed lateral angles; dorsal surface smooth or weakly veined towards the base; sessile. Coastal and inland in saline or weedy habitats. Lower leaves narrowly triangular, length about twice the width, base cuneate (basal angle less than 150°); axillary bracteoles frequent, morphologically different from terminal ones, thin, at least some of them foliaceous; margins with pointed, strongly developed lateral angles; dorsal surface strongly reticulate-veined towards the base; stalked (stalks 5-25 mm long). Restricted to estuarine salt marshes in tall vegetation?).A. Longipes DESCRIPTIONS ATRIPLEX Section TEUTLIOPSIS Dumort., Fl. Stems green with whitish, stramineous or red stripes. Venation normal dicotyledonous type.

Flowers monoecious, the pistillate ones all bracteolate and lacking a perianth. Bracteoles united at the base or up to the middle at most, not becoming cartilaginous in fruit. Seeds exclusively vertical. Prostrata group. In the British Isles this group is represented by: A.

Prostrata, A. Glabriuscula, A. Longipes, and A. The species are morphologically similar and interfertile in varying degrees TAXONOMY 0¥ ATRIPLEX IN THE BRITISH ISLES 191 and, except for A. Prostrata, all are restricted to littoral or estuarine habitats.

Many coastal populations are made up of hybrid derivatives variously intermediate between two or more species. PROSTRATA Boucher ex DC, in Lamarck & De Candolle, Fl. Francaise, 3rd ed., p. Lectotype: 'Env.

Du Havre', h. 386, marked '^A. Prostrata Boucher' in herb. Gustafsson in Opera Botanica, 39:21 (1976). Prostrata Boucher, Extrait de la Flore Abbeville et du department de la Somme, p. Triangularis Willdenow, Sp. PI., 4:963 (1806).

(Lectotype: sheet number 3, 'Ipse legi 1804 in Lido di Venezia' initialled 'W' and marked 'Atr. Triangularis', in herb. Willdenow (B), fide Taschereau in Can. Bot., 50:1583 (1972)).

Oppositifolia DC., Rapports sur les voyages botaniques etagronomiques, p. (Lectotype: specimen h. 390 marked 'Atriplex oppositifolia' in herb. Gustafsson in Opera Botanica. 39:23 (1976)). Latifolia Wahlenberg in Svensk Botanik, 9:628 (1824).

(Lectotype: Drawing No. 628, in Svensk Botanik, 9:628 (1824), fide M. Gustafsson in Opera Botanica, 39:23 (1976)).

Deltoidea Babington, Primitiae Florae Sarnicae, p. (Lectotype: Guernsey, Fort George, 1837, C.C. Babington (CGE), fide M. Gustafsson in Opera Botanica, 39:23 (1976)).

Hastata sensu Aellen in Fl. Europaea, 1:97 (1964), and sensu auct.

The species that Linnaeus called A. Hastata is the plant presently called A. Calotheca (Raf.) Fries.

(Lectotype: sheet 122L17 marked 'hastata' (LINN), fide Taschereau in Can. Fiof., 50:1585 (1972)). Plants 10-100 cm, erect, ascending, decumbent or procumbent.

Stems striate, subangular to angular, green and stramineous striped or ± reddish. Branches opposite or sub-opposite up to about two-thirds from the base.

Foliage green or reddish, non-succulent; mature leaves finely farinose or glabrous; juvenile and upper leaves glabrous to finely farinose above, grey-farinose to densely white-lepidote below. Lower leaves 2-11 cm long, 2-10 cm wide, triangular with a pair of obtuse out-pointing basal lobes; margins entire, dentate or irregularly toothed; apex acute to obtuse; base truncate to subcordate or broadly obtuse. Upper leaves smaller, triangular or lanceolate, with or without basal lobes; margins entire or toothed. Inflorescence 2-9 cm long, spiciform, composed of contiguous or irregularly spaced glomerules, terminal on stems and branches and on short stems from the axils of upper leaves, leafless except at the base. Bracteoles 2-6 mm long, sessile, triangular to triangular-ovate; apex broadly acute; base truncate to obtuse; margins entire or dentate, united at the base, lateral angles rounded and not strongly developed; herbaceous and thin or ± thickened by the presence of spongy tissue; dorsal surface smooth or tuberculate, venation obscure or prominent. Two seed types present and distinct. Brown seeds 1.5-3.0 mm wide, orbicular; radicle sub-basal, obliquely up-pointing to out-pointing.

Black seeds 1.0-2.5 mm; radicle basal, out-pointing. HABITAT AND DISTRIBUTION Halophyte, ruderal and anthropophile, common in silt, sand and shingle on sea beaches and in salt marshes around the coast except in northern Scotland (Fig. 4, omitting unconfirmed, earlier northern records). It is a characteristic component of all inland salt marsh vegetation.

The ruderal and anthropophilic biotypes, morphologically indistinguishable from the halophytic variants, are transient colonizers of freshly disturbed soil. They occur with patula along roadsides and edges of walkways and in waste ground by rubbish tips and on demolished building sites. In northern Scotland, Orkney and Shetland, A. Prostrata has frequently been confused with hybrid derivatives between other members of the.4. Prostrata group. The species reaches its northern limits in Scotland in approximately the same latitudes as A. On the east coast, the most northerly record is 3 km north-east of Dingwall, E.

On the west coast, the most northerly record is from Dumbuck, Dunbarton, v.c. Two collections from disturbed ground on Fair Isle, Shetland, v.c. 112, are introductions of the ruderal biotype of the species and do not represent extensions of its natural range. The absence oi A. Prostrata further northward is coupled with the frequent presence of two hybrids: A.

Longipes x A. Prostrata and A. Longipes x A. TASCHEREAU REPRODUCTIVE BIOLOGY Facultatively autogamous and primarily wind pollinated but also visited by syrphid flies that feed on the pollen.

Flowering August to September; seed set September to November. The staminate and pistillate flowers, closely clustered together in tight glomerules, mature and open almost simultaneously. This greatly lessens the opportunity for cross pollination. The seed, especially of the ruderal biotypes, consists mostly of the small black type. Unlike the small black seeds of A. Patula these seeds show no differences in germination response. Prostrata, the small black seeds germinate at the same rate and to the same extent as the large brown seeds.

GLABRIUSCULA Edmondston, Fl. Neotype: Keiss, near Wick, Caithness, 19. Wedgewood (K), fide M. Gustafsson in Opera Botanica, 39:15 (1976). Excellent topotype material is: sheet 739, Baltasound, Unst, Shetland, 27.

Beeby (SLBl). Babingtonii V^oods, Tourist's Fl., p. (Neotype: England, Isle of Wight, shore between Springfield and Nettlestone near Ryde, 26. 1842, the left hand specimen (K).

The description is probably based on a form of A. Rosea sensu Babington Trans. Edinb., 1: 13 (1841), fide M. Gustafsson in Opera Botanica, 39: 15 (1976)). Glabriuscula var.

Pseudocalotheca Benneti, Trans. Edinb., 20: 1 (1928). (Topotype material in herb. Arthur Bennett (BM)). TAXONOMY OF ATRIPLEX IN THE BRITISH ISLES 193 Figure 4. Distribution of Atriplex prostrata in the British Isles.

Plants 20-90 cm, prostrate, decumbent or less commonly erect. Stems striate, angular, green and stramineous striped, stout, tough-herbaceous.

Branches opposite only at the base, swollen at attachment to the main stem. Foliage green, frequently succulent; mature lower and upper leaves glabrous or sparsely fine-farinose about the base of the main veins above and sparsely farinose below; the most juvenile leaves densely farinose above and below. Lower leaves 2-7 cm long, 2-6 cm wide, triangular with a pair of out-pointing to upcurving basal lobes; margins sinuate-dentate, irregularly toothed or rarely almost entire; apex acute to obtuse; base obtuse to truncate or rarely subcordate. Upper leaves smaller, outline highly variable, lanceolate or triangular, with or without basal lobes; margins entire or toothed.

Inflorescence 2-15 cm long, spiciform, composed of loose, irregularly-spaced glomerules, terminal on upper stems and branches and from the axils of upper leaves, frequently with much reduced leaves subtending the glomerules up to about two-thirds from It 194 P. TASCHEREAU Figure 5. Atriplex glabriuscula. Bracteoles 4-10 mm long, sessile, rhombic; apex broadly acute; base broadly cuneate to obtuse or rounded; margins entire or dentate, united up to the middle; lateral angles rounded and not strongly developed; much thickened especially at the base by the presence of spongy tissue; dorsal surface tuberculate, muricate or smooth, venation obscure. Seeds 2.0-4.0 mm wide, mostly dark brown to black (rarely two types present), dull, smooth, ovate to orbicular, flattened or irregularly biconvex; radicle median and usually strongly up-pointing. HABITAT AND DISTRIBUTION Obligate halophyte confined to the littoral zone of more or less exposed sand or shingle coastal beaches.

Records of this species occurring inland (Chapman 1960; Perring & Walters 1962; Lee 1975) are erroneous. Because of the frequency with which this species has been confused with hybrid derivatives between other members of the A. Prostrata group, all distribution records for it in the British Isles need to be re-examined.

A partial distribution, based on my field work and the B. S.B.I, survey, is given in Fig. 6, but the species is much more common than the number of distribution dots would suggest. Atriplex glabriuscula is absent from the coasts of Yorkshire, Lincolnshire and Norfolk (v.cc. 61, 54, 53, 28, 27).

The Yorkshire coast needs further investigation. The Lincolnshire coast is affected by severe habitat disturbance: intensive human recreational use, military use, extensive land reclamation, sea defence wall-building, cattle grazing and intensive rabbit grazing.

TAXONOMY 0¥ ATRIP LEX IN THE BRITISH ISLES 195 Figure 6. Distribution of Atriplex.glabriuscula in the British Isles. The northern Norfolk coast is also affected by reclamation, sea-wall defence and drainage, but other ecological factors may be influencing the distribution of A. Glabriuscula here. Adam (1978) confirmed the distinctiveness of the northern Norfolk salt marshes. He noted that they show vegetational and floristic links with Mediterranean salt marshes.

The ecological niche occupied by A. Glabriuscula in other regions is here occupied by hybrid derivatives between A.

Longipes and A. REPRODUCTIVE BIOLOGY Facultatively autogamous and primarily wind pollinated.

Insects observed feeding on A. Glabriuscula and crawling over the flowers probably play only a very minor role as pollinators in the 196 P. TASCHEREAU windy habitat where this species grows.

Flowering July to August; seed set September to October. The staminate and pistillate flowers, closely clustered together at anthesis, mature and open almost simultaneously. This lessens the opportunity for cross pollination. Glabriuscula is well adapted to dispersal by sea. The thick firm bracteoles, united up to about the middle, retain the seed while it is in the water. The large air-filled cells of spongy tissue at the base of the bracteoles give them high buoyancy.

Under laboratory conditions 71% of the bracteoles examined were still floating after 24 days in continuously agitated sea water (Gustafsson 1973a). The seeds exhibit a marked dormancy.

Those planted in the greenhouse will not germinate unless exposed for some time to the fluctuating weather conditions outdoors (Taschereau 1972, 1979). In the laboratory, the seeds can be induced to germinate within nine days by exposing them to alternating daily temperatures of 10°C and 30°C (Ignaciuk & Lee 1980). Seed germination on the north-western coast of England begins during the last week of April, four weeks after the equinoctial spring tides. The main germination occurs in May with a small amount in June. Ignaciuk & Lee (1980), who observed germination on the English coast, pointed out that the alternating temperature requirement of A.

Glabriuscula seeds delays germination until after the equinoctial tides, the period of greatest environmental instability. The amplitude of the diurnal temperature cycle, noted these authors, decreases rapidly with increasing sand depth. (In measurements by Ignaciuk it was halved with each 9 cm increase in depth). Thus, the Figure 7. A triplex longipes. TAXONOMY 0¥ ATRIP LEX IN THE BRITISH ISLES 197 alternating temperature requirement also prevents the seed with its limited perisperm from germinating beyond depths greater than it can overcome. LONGIPES Drejer, Fl.

Hafniensis, p. Lectotype: Denmark, Copenhagen, Flaskekroen, sheet L92/74 no.

1, Drejer (C), fide Jones in Watsonia, 10: 250 (1975). Prostrata Boucher ex DC. Longipes (Drejer) Meijden in Gorteria, 11: 119 (1982). Plants 20-90 cm, erect or spreading, stems striate, subangular, green and stramineous striped. Branches opposite only at the base or rarely higher in large specimens. Foliage green at maturity becoming yellow with senescence, succulent; mature and juvenile leaves glabrous. Lower leaves 4-6 Atriplex longipes CHANNEL rSCAfOS PLOTTED ON OU UTM GRID Figure 8.

Distribution of Atriplex longipes in the British Isles. TASCHEREAU cm long, 3-5 cm wide, narrowly triangular with a pair of out-pointing or forward-curving basal lobes; margins entire or irregularly toothed; apex acute; base cuneate.

Upper leaves lanceolate to linear, without basal lobes or with one small lobe or a pair of weakly developed lobes; margins entire. Inflorescence 10-15 cm long, spiciform, composed of loose irregularly-spaced glomerules, terminal and in the axils of upper leaves and branches, frequently with much reduced leaves subtending the glomerules up to about two-thirds from the inflorescence base. Bracteoles consisting of two forms on the same plant: a smaller, shortly stalked to sessile, thin-herbaceous but non-foliaceous form occurring mostly in the terminal parts of the inflorescence; a larger, long-stalked, thin-herbaceous and frequently foliaceous form occurring in the mid to lower region of the inflorescence and particularly in the axils of the upper leaves and branches. Smaller bracteoles 5-10 mm long on stalks 0.5-1.0 mm long or sessile, rhombic or elongate-triangular; apex acute; base broadly obtuse or cuneate; margins mostly entire, united only at the base; lateral angles pointed, not strongly developed, dorsal surface mostly smooth, venation obscure or prominent. Larger bracteoles 10-25 mm long on stalks 5-25 (-30) mm long, ovate-lanceolate; apex acute; base cuneate to broadly obtuse; margins entire or with a few teeth, united only at the base; lateral angles pointed and strongly developed; dorsal surface smooth or slightly muricate; venation pronounced, forming a reticulate pattern towards the base. Two seed types present. Brown seeds 2.0-3.0 (-3.5) mm wide, orbicular; radicle basal to sub-basal, out-pointing.

Black seeds 1.5-2.0 mm wide, orbicular; radicle basal, out- pointing. Atriplex praecox. 4y 4 ^ ^ TAXONOMY 0¥ ATRIP LEX IN THE BRITISH ISLES 199 HABITAT AND DISTRIBUTION Obligate halophyte confined to tall salt marsh vegetation bordering estuaries. Longipes grows on a silty substratum in relatively undisturbed sites flooded with brackish water during the highest tides.

It is associated with Aster tripolium in stands dominated by Juncus maritimus and at the margins of Phragmites australis stands. In Britain, it is widely distributed, occurring in Kirkcudbrightshire, Norfolk and Cornwall, but always it is local, usually rare and commonly associated with /I. Prostrata hybrids that make it difficult to detect. Distribution in the British Isles is given in Fig.

REPRODUCTIVE BIOLOGY Facultatively autogamous and primarily wind pollinated. Flowering (?) July to August. Fruiting mid-August to early September. The flowering times can only be inferred because only fruiting Figure 10. Distribution of Atriplex praecox in the British Isles. TASCHEREAU material has been observed in the British Isles.

Staminate and pistillate flowers occur together in the terminal inflorescence, but exclusively pistillate flowers occur singly or a few together in the upper stem and leaf axils. Longipes in Britain, as in Scandinavia (Gustafsson 1972), exhibits distinct protogyny.

The pistillate flowers in the leaf axils extend a pair of receptive stigmas several days before the staminate flowers open and shed their pollen. Thus, the opportunity for cross-pollination is increased. PRAECOX Hiilphers, in Lindman, Svensk Fanerogamflora, p.

Lectotype: Sweden, Uppland, Ljustero s:n, Sarso 18. Hiilphers (marked A. Praecox) (S), fide M. Gustafsson in Opera Botanica, 39: 19 (1976). Nudicaulis Boguslaw, Lesn. Zur., 1: 30 (1846).

(Type Locality: U.S.S.R., in the vicinity of Archangel). Type material inquired for unsuccessfully at LE by M. Gustafsson whom I have followed in listing this name as a synonym. Longipes Drejer subsp. Praecox (Hiilphers) Turesson in Lands Univ. Arsskr., N.F. 2, 21(4): 6 (1925).

Plants 3-10 (-15) cm, erect or procumbent. Stems terete or sub-angular, green or red. Branches opposite up to about two-thirds from the base, the lowermost ones often long-spreading and sometimes longer than the central axis. Foliage bluish-green, often reddish tinged, succulent; mature leaves glabrous, juvenile and upper leaves finely farinose. Lower leaves 1.0-3.0 cm long, 0.5-1.3 cm wide, ovate or lanceolate with a pair of short, out-pointing basal lobes; margins entire or with a few short teeth; apex acute or obtuse; base cuneate to attenuate. Upper leaves smaller, lanceolate to linear, without basal lobes; margins entire. Inflorescence entirely axillary or also terminal, 1-3 cm long, composed of loose irregularly spaced glomerules, leafy throughout.

Bracteoles 3-5 mm long, sessile or with stalks 0.5-1.5 mm long, rhombic-ovate or triangular-ovate; apex acute or acuminate; base cuneate, obtuse to truncate; margins entire, united at the base; lateral angles rounded, not developed or slightly unilaterally developed; thin-herbaceous or membranous; dorsal surface smooth, venation obscure. Seeds 1.5-3.0 mm wide, ovoid or sub-orbicular, 0.1-0.4 mm longer than wide, not distinctly dimorphic, black or dark brown, biconvex, lustrous, smooth or patterned; radicle sub-basal, obliquely up-pointing to out-pointing. HABITAT AND DISTRIBUTION Obligate halophyte restricted to the margins of semi-protected sea inlets in northern coastal habitats. It occurs in shingle or sand in the low beach zone below the Cakile-Atriplex association, barely above the high water fucoid zone, in a region devoid of other terrestrial species. The most commonly reported habitat is close to the salt water in shingle bordering sea lochs. Praecox plants frequently form a distinctive red zone of very low, sparse vegetation clearly discontinuous from the strand plants of the middle beach. Distribution in the British Isles is shown in Fig.

REPRODUCTIVE BIOLOGY Facultatively autogamous and primarily wind pollinated. Flowering June to July; seed set August to September. The species is protogynous, the pistillate flowers extending receptive stigmas several days before the staminate flowers open and shed their pollen. Praecox matures earlier than most other Atriplex species in Britain and disappears from its habitat before October. Reproductive isolation is probably facilitated to some extent by its earlier flowering time and by the specialized ecological niche which it occupies. LITTORALIS L., Sp.

Lectotype: L. Royen (L), fide Taschereau Can. Bot., 50: 1581 (1972). Serrata Hudson, Fl.

(Lectotype: drawing of Atriplex angustifolia dentata in the upper right hand corner of the page, t.7f.4. Petiver's Herbarij Britannici (1712-15), cited by Hudson). Marina L., Mant. Littoralis var. Littoralis A. Gray, Man., 5th ed., p. 409 (1867) pro parte.

TAXONOMY OF ATRIPLEX IN THE BRITISH ISLES 201 Figure 1 1. Atriplex littoralis.

Plants 30-150 cm, erect. Stems stout, striate, sub-angular, green and stramineous striped, frequently reddish. Branches opposite only at the base. Foliage green at maturity becoming yellowish or reddish with senescence, not or slightly succulent; mature lower and upper leaves glabrous; juvenile leaves slightly scurfy. Lower leaves 2-10 cm long, 0.5-1.5 cm wide, linear to linear-oblong, without basal lobes or with a varying number of leaves possessing one lobe or a pair of out-pointing to obliquely forward-pointing basal lobes; margins entire or irregularly repand-dentate; apex acute to acuminate; base attenuate. Upper leaves reduced, linear, without basal lobes; margins usually entire.

Inflorescence long (up to 20 cm), interrupted spiciform, of densely packed glomerules widely spaced toward the base but becoming contiguous toward the apex, terminal and from the axils of upper leaves, leafless except at the base. Bracteoles 3-6 mm long, sessile, triangular-ovate; apex acute or Ungulate, frequently recurved at maturity; base cuneate to obtuse, or truncate; margins denticulate, united at the base; lateral angles rounded, not or weakly developed; usually thick- spongy; dorsal surface muricate and commonly bi-tuberculate, venation obscure. Cara Install Script Di Greasemonkey For Firefox.

Two seed types present but the black type usually more abundant. Brown seeds 2.0-2.5 mm wide, orbicular or transversely elliptic, radicle sub-basal, obliquely up-pointing to out-pointing. Black seeds 1.3-2.0 mm wide, orbicular or transversely elliptic, radicle sub-basal, out-pointing to obliquely up-pointing. TASCHEREAU Atriplex littoralis L. • 1975 onwards O pre- 1975 Figure 12.

Distribution of Atriplex littoralis in the British Isles. HABITAT AND DISTRIBUTION Obligate halophyte confined to coastal habitats. Littoralis is frequent in silt at the mouths of estuaries, in sand on more or less sheltered beaches and as a constituent of coastal salt marsh vegetation.

Prostrata, it is frequently an early colonizer of earthen sea walls. Occasionally, A. Littoralis is reported as a casual along roadsides inland, but such plants rarely persist more than one or two years. Distribution in the British Isles is given in Fig. Littoralis reaches its northern limits in northern Scotland.

On the east coast, the most northerly record is Balintore, E. It is absent from Loch Fleet, about 15 miles further north in East Sutherland, v. On the west coast, the most northerly record is Kilchattan Bay, Isle of Bute, v.c. No specimens were seen from the Orkney Islands nor was the county recorder for that region able to find any. The distribution records from there in Perring & Walters (1962) may be a mistake. TAXONOMY OF ATRIPLEX IN THE BRITISH ISLES 203 REPRODUCTIVE BIOLOGY Facultatively autogamous and primarily wind pollinated but also frequently visited by syrphid flies.

Flowering July to August; seed set September to October. Staminate and pistillate flowers occur together in the terminal inflorescence but the flowers in the upper leaf axils are primarily and sometimes exclusively pistillate. Protogynous, the pistillate flowers extending a pair of receptive stigmas several days before the staminate flowers open and shed their pollen. The bracteoles in some strains remain attached to the inflorescence axis rather than falling with the seed. At maturity the tips of these bracteoles become recurved exposing the seed which is shaken out by the action of the wind on the rigid woody stalk. In other strains the bracteoles and seed fall together. In cultivation, A.

Littoralis seeds are the fixsi Atriplex seeds to germinate but there is no information on the germination dates of plants in their natural habitat. The small black seeds are generally more abundant, comprising 70-80% of the seeds of most plants in Britain. The proportion ranges from 50-86 9f however, and some of the seeds categorized as brown are very dark brown and somewhat biconvex. In the laboratory, both black and brown seeds germinate readily at about the same rate within two weeks. There is no dormant period.

A sweet, sticky exudate is produced in droplets on the stems and main branches of A. Littoralis before and at flowering, but how this may be related to the reproductive biology is not known. Atriplex patula. TASCHEREAU 6.

PATULA L., Sp. Lectotype: No. 1221.19 in Herb.

Linne (LINN), fide Taschereau in Can. 50: 1574 (1972). Erecta sensu Smith, Fl. Angustifolia sensu Smith, Fl. Bracteata sensu Moss and Wilmott, in Camb. The taxon that Westerlund in Sveriges Atripl., p. 57 (1861) called /i.

Bracteata is a hybrid derivative between A. Longipes and A. (Lectotype: sheet marked 'Atriplex patula Lin. -bracteata Westerl.,' and initialed, 'C.A.W.' Plants 15-100 cm, erect, ascending or prostrate. Stems angular, green and stramineous striped. Branches opposite to sub-opposite up to about two-thirds from the base.

Foliage bright green at FiGURF. Distribution of Alriplex patula in the British Isles. TAXONOMY OF ATRIPLEX IN THE BRITISH ISLES 205 maturity, not changing with senescence, non-succulent; mature lower and upper leaves glabrous; juvenile leaves with fine, sparsely distributed farina (visible at x 12) on both sides but denser on the undersurface. Lower leaves 4-9 cm long, L5-4.5 cm wide, ovate-lanceolate with a pair of falcate, forward-pointing basal lobes or without lobes; margins irregularly serrate or entire; apex acute; base cuneate.

Upper leaves smaller, narrowly lanceolate to oblong-linear without or with basal lobes; margins entire or irregularly serrate. Inflorescence 1-6 cm long, interrupted-spiciform, composed of densely packed glomerules becoming contiguous towards the apex, terminal and axillary, leafless or with reduced leaves in the lower portions.

Bracteoles 3-7 (-20) mm long, sessile or with stalks 0.5- 4.0 mm long, rhombic or triangular-rhombic; apex acute or acuminate; base cuneate to broadly obtuse; margins entire or with a few short teeth, united up to the middle; lateral angles pointed, often strongly developed; herbaceous and thin, sometimes becoming foliaceous, spongy tissue never present; dorsal surface smooth or with few irregular short laciniate appendages, venation obscure or prominent towards the base. Two seed types present but intermediate types are frequent and some strains produce mostly small black seeds. Brown seeds 2.0-3.5 mm wide, orbicular, radicle sub- basal, obliquely up-pointing.

Black seeds 1.5-2.5 mm, orbicular; radicle sub-basal, out-pointing. HABITAT AND DISTRIBUTION Ruderal, fimicolous and anthropophilous weed of roadsides, pathways and barnyards, and a transient colonizer of freshly disturbed soil. This species frequently occurs in cities with /I. Prostratci on waste ground, at the margins of sidewalks, by rubbish tips and in disturbed soil on demolished building sites.

Patula rarely occurs in the littoral zone of coastal beaches. Along the coast it is primarily confined to the weedy ecotone between land and sea. It frequently occurs in soil on eroding coastal banks and about the nests of seagulls on coastal islands. Distribution in the British Isles is given in Fig. REPRODUCTIVE BIOLOGY Facultatively autogamous and primarily wind pollinated but also visited by syrphid flies that feed on the pollen. Flowering mid-June to October; seed set September to November.

The staminate and pistillate flowers are closely clustered together in tight glomerules. Patula exhibits a slight protogyny, the female flowers extending receptive stigmas about one to four days before the anthers open. This species is the earliest flowering Atriplex in the British Isles. In Manchester, some plants begin flowering as early as 1st June. In the laboratory, the brown seeds germinate within about two weeks but few black seeds will germinate without being scarified and subjected to a period of alternating temperatures. Section SCLEROCALYMMA Ascherson, Fl.

Brandenb., p. Stems whitish, pale brown or reddish. Venation kranztypus. Flowers monoecious, the pistillate ones all bracteolate and lacking a perianth.

Bracteoles united up to the middle becoming cartilaginous in fruit. Seeds exclusively vertical.

Lectotype: in Hortus Siccus Cliffortianus (BM), fide Taschereau in Can. Bot., 50: 1591 (1972).

Arenaria Woods in Phytologist, 3: 593 (1849), non Nuttall. Maritima Hallier in Bot. 10 (1863), non Crantz nec Pallas. Sabulosa Rouy in Bull. Fr.,31: 20 (1890). Plants 6-30 cm, decumbent with wide-spreading, ascending lower branches. Stems smooth or sub- angular, reddish or yellowish, more or less sparsely scaly, straight; much branched in a diffuse asymmetrical pattern.

Lowermost one to three pairs of branches opposite or sub-opposite, the remainder alternate. Foliage whitish-green or greyish-green, non-succulent; mature and juvenile leaves covered with a fine scaly layer on both surfaces but more densely covered below. Lower leaves 1.5^.0 cm long, ovate to lance-ovate, sinuate-dentate with larger basal lobes; base cuneate to a short petiole; apex obtuse. Upper leaves smaller, narrower, lanceolate or oblong, sinuate-dentate or entire, obtuse, mucronate. Inflorescence glomerulate in upper leaf axils. Bracteoles 6-7 mm long.

TASCHEREAU Figure 15. Airiplex laciniata. Sessile or with short stalks, broadly rhombic, whitish-green becoming scaly-black with maturity; apex acute; base cuneate; margins entire or with a few short teeth, united up to the middle, with lateral angles obtuse and strongly produced; cartilaginous, becoming thickened and hardened in the lower half, spongy tissue never present; dorsal surface smooth or with irregular pointed or flattened and wing-like tubercles or projections in the lower half, venation ± prominent. One seed type present: light brown, 3.5-4.0 mm wide, transverse-elliptic, dull, smooth; radicle median out- pointing to ascending, thick and prominent with apex strongly produced. HABITAT AND DISTRIBUTION Coastal halophyte of sand or sand and cobble beaches. Widespread in the British Isles but often local.

Where it occurs, it is usually present in low numbers as more or less widely-scattered individuals and after some years absence it may reoccur in a former locality. Observed in abundance only at Dunnet Bay, Caithness, Scotland. Distribution in the British Isles is given in Fig. REPRODUCTIVE BIOLOGY Facultatively autogamous and primarily wind pollinated. Isolated plants grown in the greenhouse produced normal amounts of viable seed (Taschereau unpublished). 3d Railroad Concept And Design Free Download. Flowering August to September; seed set September to October. Laciniata is well adapted to dispersal by sea.

The cartilaginous bracteoles, united up to the middle, tenaciously retain the seed and soften little after soaking in sea water for several days. In the laboratory, bracteoles of this species floated in sea water for up to ten days (Taschereau 1970). Ignaciuk & Lee (1980) immersed bracteoles in constantly agitated salt water (600 mM NaCl). After five days, 20% of the bracteoles remained floating, but by eight days all had sunk. Ignaciuk & Lee (1980) found that seeds immersed in salt water (600 mM TAXONOMY 0¥ ATRIPLEX IN THE BRITISH ISLES 207 Fic.URi; 16.

Distribution Airiplcx laciniata in the British Isles. NaCl) tor up to 30 days remained viable. Furthermore, these authors report that the seeds could germinate at this salt concentration and produce healthy seedlings although the growth rate of the plants was reduced. The seeds exhibit a marked dormancy. Those planted in the greenhouse will not germinate unless exposed for some time to fluctuating weather conditions outdoors (Taschereau 1970). In the laboratory, however, seeds readily germinate after being exposed to a period of alternating daily temperatures (Ignaciuk & Lee 1980).

As in the case of A. Noted Ignaciuk & Lee ( 1980), the alternating temperature requirement of A. Laciniata delays germination until after the spring equinoctial tides, the period of greatest environmental instability.

This mechanism, they observed, also prevents the seed with its limited perisperm reserve from germinating beyond depths greater than it can overcome. TASCHEREAU HYBRIDS A. Glabriuscula x longipes Hybrid derivatives between /I. Glabriuscula and A longipes aie frequent on exposed coastal beaches in northern England and Scotland. They occur with A.

Gl.briuscula in the habitat characteristic of that species in regions where A. Longipes is absent. Glabriuscula x praecox Rare in northern Scotland and Shetland. Occurring in the same habitat as the parent species.

Glabriuscula x prostrata Rare, from southern England to south-western Scotland and on the eastern coast of England. Occurring in the same habitat as the parent species. Earlier literature reports (Moss & Wilmott 1914; Jones 1975b) of its frequency are unconfirmed. Longipes x prostrata Frequent wherever the parent species occur together; occasional inland in salt marshes and waste places.

Hybrid derivatives between A. Longipes and A.

Prostrata are common in sand and shingle on exposed coastal beaches in all regions of the British Isles, frequently in areas remote from one or both parent species. Gustafsson's (1973b) experimental work with this hybrid in Sweden showed that it is present there as a more or less well-stabilized variant. Comparable plants are frequent in northern Scotland and occasional in Shetland where they occur on exposed coastal beaches, a habitat colonized by neither of the parent species. Occasionally found with A. Praecox on the shores of somewhat less exposed sea inlets. Littoralis x prostrata Occasional on the eastern and western coasts of England in disturbed habitats where both parents are present in abundance. Littoralis x patula Known from only one locality in Midlothian, v.c.

83, where both parents were present in disturbed waste ground by the coast. ACKNOWLEDGMENTS Mats Gustafsson's work on the A. Prostrata group in Scandinavia has been fundamental to my understanding of Atriplex in the British Isles. I am deeply indebted to him for his personal assistance, many stimulating conversations and for his warm hospitality during my visits to Sweden. The Atriplex Survey undertaken by members of the Botanical Society of the British Isles in co- operation with the Biological Records Centre, Monks Wood, further extended the geographic area covered by this study.

To the Centre and to the B. S.B.I, members who took such care in sending me specimens and data, I am grateful. My discussions and field work with Richard P. Libbey, who took me to Atriplex sites in Norfolk, were enjoyable and helpful and I benefited also from my visits to and discussions with two former students of the genus, Barbara Hulme and Elizabeth Jones. The first three years of this study were financed by a scholarship from the National Research Council of Canada. The continuation of the study for a further two years was made possible by the generous financial help I received from John Meredith and the late Hilda Meredith.

I acknowledge the help and encouragement of Professor D. Valentine, my supervisor during the course of this study. The drawings were prepared by B. TAXONOMY OF ATRIPLEX IN THE BRITISH ISLES 209 REFERENCES Adam.

Geographical variation in British saltmarsh vegetation. In Rechinger, K. Hegi Illusirierte Flora von Mitteleuropa, 2nd ed.

Flora Europaea. Memoirs, journal and botanical correspondence of Charles Cardale Babington. Monograph of the British Atripliceae.

Edinb., 1: 1-17. Manual of British Botany. Handbook of the British flora. Revised by J.

Salt marshes and salt deserts of the world. Interspecific relationships and intraspecific variation of Chenopodium album in Britain.

The taxonomic delimitation of the species. The Betacyanins a class of red pigments in the Centrospermae. Recent developments in the chemistry of natural compounds, pp. British Plant List.

Manual of the botany of the northern United States. Distribution and effects of paracentric inversions in populations oi Atriplex longipes. Hereditas, 71: 173-194. Evolutionary trends in the Atriplex triangularis group of Scandinavia.

Hybrid sterility and chromosomal differentiation. 126: 345-392.

GusTAFSSON, M. Evolutionary trends in the Atriplex triangularis group of Scandinavia. Spontaneous hybridization in relation to reproductive isolation. Notiser, 126: 398-416.

Evolutionary trends in the Atriplex triangularis group of Scandinavia. The effects of population size and introgression on chromosomal differentiation. 127: 125-148. GusTAFSSON, M. Evolutionary trends in the Atriplex prostrata group of Scandinavia, 4. Taxonomy and morphological variation. Biological Records Centre instructions for recorders, p.

Index herbariorum. Studies on some British species o/ Atriplex. University of Edinburgh. Artificial hybrids in the genus /1/r/p/f.v. The germination of four annual strand-line species.

New Phytol., 84: 581-591. Taxonomic studies of the genus /Irr/p/t'.x (Chenopodiaceae) in Britain.

Hybridization and the flora of the British Isles, pp. British and Irish herbaria. The conservation of British inland salt marshes. Van der (1970). Biosystematic notes on Atriplex patula L.

Littoralis L! The Cambridge British flora. Plates 172-188. Atlas of the British flora. Introduction, in Asberg.

Translation of Linnaeus's Oland and Gotland Journey 1741. The genus Mx' p Q in Nova Scotia: taxonomy, ecology and distribution. Acadia University.

Taxonomv and distribution o{ Atriplex species in Nova Scotia. 50: 1571-1594. Atriplex praecox Hiilphers: a species new to the British Isles.

Taxonomy of the genus Atriplex //; Great Britain. University of Manchester. The cause of placiotropv in maritime shore plants.

The species and the variety as ecological units. The genotypical response of the plant species to the habitat. Studies in the genus Atriplex.

Flora ol the British Isles. {Accepted March I9cS4) Watsonia, 15, 21 1-219 ( 1985) 211 Field studies, cultivation experiments and the taxonomy of Atriplex longipes Drejer in the British Isles p. TASCHEREAU' Department of Botany, University of Manchester ABSTRACT Atriplex longipes Drejer (Chenopodiaceae). A rare and elusive species known in the British Isles only since 1977, is reported from eight localities. Its rarity in Britain is due to the specialized habitat it occupies and the facility with which it hybridizes with A. Hybrids with A. Prostrata are frequent in all localities and cultivation studies reveal that hybrid plants can be morphologically indistinguishable from A.

Longipes s.str. Taxonomic characters separating /I.

Longipes from y4. Prostrata are given. The presence of stalked bracteoles is not sufficient to distinguish /I. Longipes from other taxa. Field studies indicate that hybrid derivatives between A. Longipes and A. Prostrata and between /I.

Longipes and /I. Glahriuscula are occasional to frequent on most coasts of the British Isles. For morphological, genetic, ecological and practical reasons, /I.

Longipes is maintained at the species level. INTRODUCTION Atriplex longipes Drejer is a member of the A. Prostrata group {Hastata complex). The group comprises a number of partially interfertile and morphologically similar taxa found on the coasts of western Europe and elsewhere. In the British Isles it is represented also by A. Prostrata Boucher ex DC.

Hastata auct.), A. Glahriuscula Edmondston and A. Praecox Hiilphers. Aellen (1964) reported that Atriplex longipes was 'recently found to be widespread in the British Isles...' .He noted that it was often confused with A. Prostrata and A.

Glahriuscula and that 'in Britain it has commonly been called A. Hracteosa Aellen based his report on data provided by B. Jones (1975a) reported on plants she said were 'best described as variants in the A. Hastata complex which resemble A. Longipes^ She concluded that 'a few plants very similar to this species have been found in Britain', but that they were 'less common in this country than suggested by Hulme'. Gustafsson (1976) noted that A. Longipes had been reported from the British Isles, but stated that all the material he had seen from there could be referred to other species.

My search of the major British and Irish herbaria failed to reveal a single specimen of A. The herbarium of B. Hulme contained none and Jones's (1975) specimen from Brean, Somerset (OXF!) proved to be a hybrid derivative between A. Longipes and A.

In 1977, I reported the occurrence of A. Longipes at Wigtown Bay, v.c. 73, thus confirming the presence of this species in the British Isles.

The British specimens compared well with Scandinavian material in C, LD and S and my determination was confirmed by M. Gustafsson (Taschereau 1977). More specimens were later collected in Norfolk (Libbey 1977) and further field work revealed another locality in south-western England.

Despite this, A. Longipes remains an elusive and exceedingly rare species in Britain, persisting only in relatively undisturbed tall estuarine saltmarsh vegetation.

Longipes has a mainly Scandinavian distribution and. Until 1977, its occurrence outside of this area was unconfirmed.

The species has since been reported from Holland and more recently (Garve 1982) from Germany. Gustafsson's (1972, 1973a, 1973b, 1-974, 1976) experimental studies of A. Longipes and its relationship to other members of the A. Prostrata group provide the basis for understanding the biology of this species in Britain and elsewhere. *Present address: Institute for Resource and Environmental Studies, Dalhousie University, Halifax, 'ova Scotia, Canada, B3H 3E2 212 P. TASCHEREAU The present paper reports the results of field and cultivation work with.4. It describes the important taxonomic characters of this species and compares them with those oiA.

It gives data on the specialized habitat occupied by A. Longipes in the British Isles and discusses the frequency of hybridization between A. Longipes and A. Prostrata here.

It gives reasons why A. Longipes is appropriately maintained at the species level and why this taxon is the key to understanding much of the variation in the coastal members of the A. Prostrata group in the British Isles. MATERIALS AND METHODS FIELD STUDIES Extensive field studies were made between 1974 and 1978 as part of a broader study of the taxonomy of the genus Atriplex in the British Isles (Taschereau 1979). In 1977, this field work was supplemented by a network survey of the coastal Atriplex species undertaken through the Botanical Society of the British Isles in co-operation with the Biological Records Centre, Monks Wood Experimental Station.

In this way, specimens and data on Atriplex species, hybrids, and their habitats were obtained from areas within all the major plant regions of Britain (Heath & Scott 1974) and all coastal regions except the Outer Hebrides and the coasts of Ireland. CULTIVATION WORK Seeds were taken from typical herbarium specimens of A. Longipes collected from three, widely- separated, British populations: Palnure, Kirkcudbrights., v.c. 73; Brancaster, W. Norfolk, v.c.

28; and Penpoll, E. Cornwall, v.c. The seeds were separately sown in the autumn in trays of John Innes Seed Compost and placed in an unheated greenhouse. The seedlings were later transferred to individual pots containing John Innes Potting Compost and repotted as necessary. A number of seedlings were treated in various ways (Table 1) to investigate the influence of shading, pruning and crowding on bracteole morphology and bracteole stalk development. The plants used in these experiments were grown from seed taken from a single specimen (Brancaster 76-18) that had numerous stalked bracteoles (larger stalks 10-15 mm long). MEASUREMENTS AND HERBARIUM STUDIES The following taxonomic characters were examined in both the wild plants and their cultivated offspring: bracteole stalk length, plant height, leaf base angle, lower leaf morphology and the ratio of black:brown seeds produced.

(These characters are discussed in detail in Taschereau (1985a)). Evidence of hybridization was provided by progeny tests.

Material from the following herbaria was examined: ABD, BM, C, CGE, DBN, E, K, LD, LIV, LIVU, MANCH, NMW, OXF, S, SLBI, TCD (abbreviations are according to Kent & Allen ( 1984), and Holmgren et al. My specimens are deposited in MANCH. RESULTS HABITAT In the British Isles, Atriplex longipes is restricted to a very specialized habitat: relatively undisturbed, tall, salt marsh vegetation in the upper zones of estuaries on a silty substratum flooded with brackish water during the high spring tides.

It is associated with Aster tripolium in stands dominated by Juncus maritimus and present at the margins of Phragmites australis stands. MORPHOLOGY IN THE WILD The plants mature by September and the aerial parts then disappear from the habitat. During seed maturation, the lower leaves change from succulent to thin, turn from green to yellow, and drop off. This process begins in the lowermost parts of the plant and progresses up to the top. In mature leaves separated from the plant, autolysis is very rapid.

Rotting begins in the apical parts of the leaf and proceeds toward the central and basal parts. Each bracteole pair has a very small point of attachment to the stem. At maturity the bracteoles readily break at this point separating from the plant and falling whenever the plant is disturbed. ATRIPLEX LONGIPES IN THE BRITISH ISLES 213 TABLE 1. THE INFLUENCE OF SHADING, PRUNING, CROWDING, ON BRACTEOLE MORPHOLOGY AND BRACTEOLE STALK DEVELOPMENT IN A. LONGIPES Treatment Controls Influence of treatment SHADING Seedlings planted outdoors under dense growth of closely cultivated tall herbaceous annuals Seedlings planted outdoors, well-spaced in an open, unshaded plot adjacent to the plot with the treated plants Controls and treated plants did not differ significantly. Some plants in each group developed stalked bracteoles (stalks up to 5 mm long).

PRUNING Potted plants in greenhouse treated in the following way: Pruned at different developmental stages: (1) Before flowering (2) Immediately after flowering (3) During early fruit maturation Different parts pruned: (1) Central axis one-third up from the base (2) Side branches to various degrees Potted plants unpruned, interspersed with the pruned plants on the same greenhouse bench Controls and treated plants did not differsignificantly. Some plants in each group developed stalked bracteoles (stalks up to 5 mm long). CROWDING From 10 to 20 seedlings were planted in J.I. Compost in 14cm pots and allowed to grow to maturity without being repotted Seedlings potted separately and repotted regularly into fresh J.I. Compost every 2-3 weeks Treated plants differed markedly from the controls and plants in other treatments: (1) Treated plants were approximately half the height of the controls (2) Bracteoles were morpholog- ically similar to those on the parent plant (i.e.

Thin, foliose, reticulate-veined) (3) Stalks up to 10 mm long developed on numerous bracteoles Table 2 compares the taxonomic characters of A. Longipes with those of A. Prostrata, the species with which it most frequently hybridizes. In the field, two morphological variants were observed. These differ in size and habit: one is 1 m or more high with long (up to 50 cm) opposite lower branches and a straggling growth habit. It occurs at the margins of Phragmites aiistralis stands. The other variant is 30-80 cm high with short (10-20 cm) alternate branches and an erect habit.

It occurs in dense stands of J uncus maritimus. The influence of environment on the development of these two variants remains uncertain but observations on their cultivated offspring revealed that hybridization with A. Prostrata was involved in all the larger ones. MORPHOLOGY AND CULTIVATION The cultivation results are summarized in Tables 1 & 3.

The most important findings are that development of the bracteole stalk on A. Longipes is subject to environmental modification, and that hybrids involving A. Prostrata can be morphologically indistinguishable from A. Stalk development on the bracteoles of A. Longipes was affected by cultivation conditions. All the cultivated offspring had shorter stalks than their wild parents. PenpoU (77-43) plants, discussed above, showed no clear evidence of hybridization with A.

Prostrata and the cultivated plants came from seed selected from herbarium specimens exhibiting extreme bracteole stalk development (stalks up to 30 mm long). Despite this, 70% of the cultivated progeny of these plants with long- 214 P. TASCHEREAU TABLE 2.

TAXONOMIC CHARACTERS OF A. LONGIPES AND A. PROSTRATA COMPARED IN BRITISH AND SCANDINAVIAN (GUSTAFSSON 1976) PLANTS MEASUREMENTS BASED ON CULTIVATED PLANTS ARE SHOWN IN BRACKETS Characters A. Prostrata LOWER LEAVES Outline Length/width ratio Base angle Elongate-triangular to rhombic 1.2- 1.5 British Isles (Cult.: 1.2-1.6) 1.3- 3.5 Scandinavia 100°-150° British Isles (Cult.: 120°-180°) 50°-145° Scandinavia Deltoid-triangular to triangular hastate 0.7-1.