An intensive study of the cytology and sexual behaviour of the classical Spartina x townsendii agg. and their putative parents in southern England reveals a more complex chromosomal situation than had previously been reported, including aneusomaty in Southampton S. alterniflora populations and in S. x townsendii agg. Chromosome numbers are 2n= 60 and 2n= 62 for S. maritima and S. alterniflora respectively and 2n= 62 and 2n= 120, 122 and 124 (chromosome races) for S. x townsendii F1 and Amphidiploid derivatives respectively. These numbers, though their summation is less precise than hitherto reported, still support the hybridity and amphidiploid origin of S. x townsendii agg. and the process is further confirmed by meiotic pairing data and the discovery of wild backcross hybrids (2n= c.90and 2n=76) near the site of origin at Southampton. The 2n= 62 chromosome number of S. alterniflora, out of line with the x= 10 base number of the genus, is explained as a polysomic condition (2n= 60+ 2) and needs further investigation with a study of North American populations. It is suggested that chromosome races found in S. x townsendii Amphidiploid will provide adaptive variation in this highly significant and successful amphidiploid.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Royal Botanic Gardens, Kew and Springer are collaborating with JSTOR to digitize, preserve and extend access to Kew Bulletin.The only chromosome study of any magnitude in the family Bromeliaceae is that of Lindchau (1933) who reported chromosome numbers for several genera, and about 50 species in all. Matsuura & Suto (1935) have also published some counts. These results suggested, by the variety of basic numbers they purported to reveal, cytological support for the taxonomic separation of the homogeneous subfamilies Pitcairnioideae and Tillandsioideae, with the somewhat more variable Bromelioideae holding an intermediate position between them. Contrary to these early results, the present investigation has revealed that the Bromeliaceae is cytologically very homogeneous throughout, with little variation in basic number or level of ploidy. This, in view of the wide morphological diversity (Mez, 1896) and ecological adaptations (Pittendrigh, 1948), is somewhat remarkable. MATERIALS AND METHODSExcept in the Pitcairnioideae, where representatives are terrestrial in habit, the epiphytic nature of Bromeliads and their consequent lack of an active root system, apart from those of seedlings, makes the usual source of somatic chromosomes unavailable, neither is it possible to resort to stem apical tissue as an alternative owing to the relatively unbranched habit in this family. Thus, apart from the few instances which will be mentioned, all chromosome counts were obtained from flower bud material at meiosis. This is a disadvantage in certain genera and especially in Billbergia Thunb. where the flower buds undergo meiosis at such an early stage of development that to obtain them would often mean destruction of the leaf rosette, a procedure which is frequently impossible in a botanic garden collection.When used, root-tips were pretreated overnight at I-20C. in distilled water or at room temperature for 4 hours in alpha-bromonaphthalene, hydrolysed Io minutes in IN HCl and stained in basic fuchsin. Anthers, used fresh or fixed 24 hours in I :3 acetic alcohol plus a drop of ferric chloride, were squashed in propionic carmine.Photographs were taken on a Zeiss photomicroscope reproduced at X o1050.Voucher specimens of all species and hybrids counted are deposited in the Kew Herbarium. Table I, pp. 162-3. Precise localities of origin of most of the species at Kew are lacking, due largely to their acquisition from other botanic gardens without any appended data. Counts are distinguished according to the material (meiotic or somatic) from which they were derived. For comparison, and where appropriate, counts of Lindchau (1933) and Diers ( 961) are also included in the list. L I6i 162 KEW BULLETIN V...

SUMMARY The history of the genus Spartina in Britain is traced from herbarium records and publications. This historical information and a morphological study of living plants strengthen the view that sterile S. x townsendii arose in c. 1870 in Southampton Water as a hybrid between native S. maritimu and introduced American S. alterngora. The fertile Amphidiploid arose in 1890 and spread at the expense of the species populations. It is shown that species and hybrids of the S. x townsendii complex can be separated by a combination of only a few well‐defined morphological characters, and by stomata and pollen grain details. A polyhaploid seedling of low vigour among Amphidiploid progeny is described and its existence and resemblance to S. x townsendii F1 emphasizea the relationship between this PI hybrid and the Amphidiploid. Putative backcross hybrids of S. x townsendii Amphidi‐ ploid with X. alternijfora in Southampton Water provide conclusive evidence of S. x

No abstract

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.Royal Botanic Gardens, Kew and Springer are collaborating with JSTOR to digitize, preserve and extend access to Kew Bulletin.