146 



enclose a bifurcate thalamus (fig. 11). The question arose : 

 is this bifurcation to be put down to subséquent rupture 

 of the growing réceptacle or to dichotomy in its youngest 

 stage of development. My spécimens — only eight of 

 them were kept — showed the following: 



1". Sepals 6 outer and one inner alternating with two 

 of the former. The réceptacle (thalamus) is split up into 

 two and even part of the peduncle is affected in the same 

 way. Drupels hâve evidently been implanted on the outer 

 surfaces of the parts of the réceptacle. The whole thing 

 makes the impression of bifurcation caused by astronger 

 growth of the periphery in contrast with the axis. 



2". Sepals 5. The cleft divides three from two sepals. 

 For the rest as in 1". 



3". Sepals 6, divided into two sets of 3. 



In none of thèse cases has the calyx been completely 

 split up showing cohésion near the top of the divided sepals. 



4". The cleft goes through the middle of two sepals in 

 such a way that on the one side there is one entire sepal, 

 on the other still two. 



5". The cleft divides three sepals from two. The pe- 

 duncle is scarcely affected, but the calyx-tube is very dis- 

 tinctly split up. 



6". Two sepals divided from three, peduncle only a 

 little affected. Two inner sepals, of which one is petaloid. 



7". Two sepals divided from three. There is an inner 

 whorl of sepals, of which one is more or less petaloid. 

 Peduncle distinctly but slightly split up. 



8". Two ordinary and one small sepal at both sides of 

 the cleft. The two small ones might be the halves of a 

 large one. Calyx-tube split up at one side only, the tops 

 of the (double) réceptacle strongly curved so as to touch 

 one another. 



From thèse eight cases it appears 



a) that the calyx has been augmented, a fact calling to 



