FIELD MEETINGS, 1951 
45 
tension shoot, fruiting shoots gathered as the liips turn colour and be- 
fore the se])als drop and a sample of the armature of the main stem. 
The method of taking a long sliver of bark from the stem to show the 
types and spacing of the prickles was demonstrated. This is piessed in 
the usual way so that it is easily mounted with the prickles in profile. 
Notes of the habit of the bush, the colour of the ]>etals and their claws 
and of the stamens, as well as of the j)erfume should be made at the 
time of collection. In glandular forms the odour given off on rubbing 
should be noted as well. The necessary specimens can all be gathered 
on two visits. 
It woidd take up too much space here to detail the characters ol 
the species as they were demonstrated on the living plants, but it may 
l)e of interest to record the identifications of the marked bushes that 
were discussed. These are given in accordance wdth Wolley-Dod’s 
monograph, together with an opinion of the true nature of the plant. 
The numbers are my field numbers, by which the specimens placed in 
the herbarium at Kew can be identified. 
50.128. Rosa cunina L. var. sevticosa (Acli.) Baker • /{. canina x arvemis. 
50.129. R. canina L. var. senllcosa (Acli.) Baker = R. canina x arvensis. 
50.130. R. 7~ubiginosa L. var. rotundifolia Rau = R. rubiginosa form. 
50.131. R. tomentosa Sm. var. typica W.-Dod = R. tomentosa x nibiginosa. 
50.132. R. rubiginosa L. var. typica W.-Dod = /?. rubiginosa typical. 
50.133. R. canina L. var. sylvularum (Rip.) Rouy = R. micrantha x arvensis. 
50.134. R. canina I,, var. ftexibilis (Desegl.) Rouy - R. canina x arvensis. 
50.135. R. tomentosa Sm. var. pseudocuspidata (Crep.) Rouy = R. tomentosa 
(doubtfully pure). 
50.136. R. canina L. 1. vinacea (Raker) Rouy = R. rubiginosa x amensis. 
•50.137. Near R. micrantha Sm. var. septicola (Desegl.) Gren. = R. micrantha x 
arvensis. 
50.138. R. canina L. f. recognita Rouy = R. canina x •? micrantha x arvensis. 
50.139. R arvensis Huds. var. vulgaris Ser. = R. arvensis typical. 
50.140. R. tomentosa Sm. var. typica W.-Dod -- R. tomentosa, doubtfully pure. 
50.141. R. tomentosa Sm. f. foetida (Bast.) Ser. = R. tomentosa x ? micrantha. 
The ]>entaploid roses are necessarily matriclinal in character, since 
the egg cells receive four sets of chromosomes from the female parent 
to one set from the pollen parent. The appearance of hybrids in which 
11. arvensis is a parent suggests that this species is almost invariably 
tlie jmllen parent, a view that is receiving confirmation from breeding 
work being carried on at the John Tnnes Horticultural Institution by 
Mr. G. D. Rowley, who was a guest at both excursions. Although the 
suggested parentages apply only to the particular hushes, it is, never- 
theless, probable that the corres])onding Wolley-Dod names are normally 
attached to similar hybrids. The mention of two species only in a sug- 
gested parentage does not exclude the possibility that genes from other 
.species are present in the inheritance of a particular bush. It com- 
monly implies that characters from the two species mentioned can be 
definitely recognised and that evidence of other species, although often 
susi)ccted, is inconclusive. Finally, it must be emphasised that the 
conclusions reached depend upon the correlation of as many characters 
