PROLIFERATION OF SPIKELETS IN BRITISH GRASSES 



55 



The literature yields no well defined name free from ambiguity which covers either 

 all the viviparous Deschampsiae, or all those other than D. alpina. 



8. Summary 



8.1. The continued growth of the spikelet axis and the bearing of foliage leaves 

 distally upon it is termed the vegetative proliferation of spikelets. If these leafy shoots 

 form deciduous plantlets, which are the regular diaspores of the plant, the phenomenon 

 is called vivipary by many authors. 



The history and confusion of these terms is traced. Proliferation is an older and 

 more accurate usage and should be employed in morphological and physiological dis- 

 cussions. Vivipary is useful in its restricted sense to describe regular propagation in 

 this manner, as it has ecological meaning. 



8.2. Five races of viviparous grasses are native in Britain ; these have an arctic-alpine range 

 about the North Atlantic. 



Festuca vivipara (L.) Sm. 

 Poa alpina L. var. vivipara L. 

 Poa X jemtlandica (Almqu.) Richt. 



Deschampsia cespitosa (L.) Beauv. var. pseudalpina (Syme) Druce. 

 Deschampsia alpina (L.) Roem. & Schult. 



Poa bulhosa L. var. vivipara Koel., a plant of dry disturbed places, is an alien in the 

 British flora. 



8.3. The names of the Poae are accepted without discussion. 



8.4. Festuca vivipara is clearly of the F. ovina agg. The vegetative variation within 

 F. vivipara is almost as wide as in F. ovina agg., thus F. vivipara cannot be placed as a 

 subspecies or variety of a segregate of F. ovina. 



F. vivipara includes both fully and partially viviparous forms, which may be 

 distinguished by Turesson's names viviparae and semiviviparae respectively. 



8.5. Deschampsia cespitosa var. pseudalpina and D. alpina are recognized. Although 

 only the latter is at all well defined, and transition forms exist between the two. Neither 

 vivipary nor the characters that distinguish the forms are completely constant. The 

 forms are not separate cytologically. 



An argument is advanced that the more proliferated var. pseudalpina is further 

 from seminiferous D. cespitosa than D. alpina. They are left separate, partly in view 

 of their differences in ecological range and partly because of the lack of a suitable well 

 known name. 



9. Acknowledgments 



It is a pleasure to acknowledge gifts of specimens from Mr. C. E. Hubbard, Prof. 

 T. J. Jenkin, Prof. T. G. B. Osborn and Prof. H. J. Venema. Thanks are due to the 

 Linnean Society and the British Museum (Natural History) for facilities to inspect 

 specimens. 



It is with gratitude that the guidance and encouragement of Prof. W. H. Pcarsall, 

 F.R.S., throughout this work is remembered. During the period of this study the writer 

 received a grant from the Agricultural Research Council. 



BIBLIOGRAl^HY 



AKERBERG, E., 1942, Cytogenetic studies in Poa pratensis and its hyhi id w ith Poa aliMiia, lUncdiLas, 28, 1. 

 ARBER, A., 1934, The Grammcac. Cambridge. 



AVDULOW, N. P., 1931, Karyo-systcmatischc Uatcisuchung dor I'amilic Gramincca, Dull. Appl. But. 

 Gen. & Plant Br. Suppi, 44. 



