4 8 
Edith R. Saunders. 
to have reproduced the deformity. 1 The number of plants raised is 
not, however, stated. Again recently, while this account was already 
in progress, a specimen was exhibited at a meeting of the Linnean 
Society by Professor Dendy, on behalf of Dr. N. C. Macnamara. 2 The 
spike in question had, according to the account, been cut from a plant 
grown from seed of a sport that had appeared in 1907, and showed the 
extreme form in which the five divisions of the corolla are represented 
by five stamens. Here too it was observed that the peculiarity came 
true from seed. In none of these cases, however, have we full 
information regarding the statistical evidence on which the general 
statement as to inheritance is based, nor have I been able to find 
any account of observations continued beyond the first generation. 
The following experiments were therefore undertaken with a view 
to ascertaining to what extent this form breeds strictly true, and 
the nature of its relation to the type. From the results which 
follow it will be seen that offspring of heptandra parents all show 
dialysis and staminody of the corolla, hut the extent to which these 
malformations are exhibited varies not only in different individuals, 
but in different flowers on the same individual ; further, that the 
heptandra form is related to the type as recessive to dominant. 
A specimen of D. purpurea, exhibiting the peculiarities above 
described was shown to me in 1906 by a friend, in whose garden in 
Cambridge it had appeared among a number of normal plants which 
had been grown from a single packet of seed. 3 It was then already 
late in the season, and none of the flowers which were artificially 
fertilised set seed. A considerable number of good capsules were, 
however, gathered from the flowers which had been exposed and 
naturally fertilised. From seed thus obtained a first generation of 
plants were flowered in 1908, and a second in 1910. The following 
account contains the results of observations made during these two 
seasons. 
I.—Description of D. purpurea heptandra. 
All individuals showing the general features previously men¬ 
tioned, viz., dialysis and staminody of the corolla, are included 
under this title. 
In heptandra, as in the type form, the axillary buds at the 
1 Ibid., 1904, 2, p. 208. 
2 See Proceedings of the Linnean Society for 1910, p. 106. 
3 Most observers have, as in this case, recorded only a single 
sport in the first instance, though Henslow’s case appears 
to have been an exception, 
