W. C. Worsdell. 
64 
flower at the other end of the row contained tri-locular, while the 
others, which are much laterally compressed, contained two-locular 
ovaries; there was also a slight reduction in the number of perianth- 
lobes, stamens and styles, many of the latter being fused. 
Fig. 18. Pistils of Campanula media shewing two stages of “ concrescence.” 
In A ovaries united, but styles free. 
A Crocus, which I observed last spring, produced a flower of 
exceptional size and interest; it was the result of “ fusion ” 1 of two 
normal flowers; but instead of twelve, there were only ten perianth- 
leaves present, this being due to the fact that two of the members 
of the outer whorl of each flower had fused respectively with a 
member of the inner whorl of the other flower; hence these two 
leaves (the products of fusion) are situated obliquely, i.e. partly in 
the inner, partly in the outer whorl. There are six stamens alter¬ 
nating as best they can with the perianth leaves. There are two 
distinct styles and sets of stigma-lobes. The ovary is double the 
ordinary size, irregular and asymmetrical in contour, and composed 
of six carpels; it shows no obvious signs of being a result of fusion 
except in the increased number of its parts. 2 (Fig. 19.) 
Allied to this case are the striking phenomena presented 
by many peloric terminal flowers, as in Linaria vulgaris, Strep- 
tocarpus, Digitalis purpurea; an instance of the latter came 
1 I beg the reader not to interpret this term in the kinetic , but 
in the purely descriptive and metaphorical sense, wherever in 
the description of the facts pure and simple I use such 
terms as 11 fusion,” “ fasciation,” cohesion,” “coalescence,” 
“ union,” they must be understood in this way. 
2 Dr. Church points out to me that in these cases of flowers we 
obtain the pure type of fasciation, devoid of spiral torsion, 
owing to the absence of secondary internodal elongation. 
