50 CARUNCLES. [ BO K i- 



seen in Euonymus, it will be easily seen that the caruncle of 

 Euphorbia is, as it were, nothing but a rudiment of the more 

 highly developed arillode of the Spindletree ; and this analogy 

 will be still more evident, when we shall have seen states of 

 the false aril intermediate between these two. If the results 

 to which I have been led, by applying the single fact of 

 M. Mirbel to a certain number of plants, are at all interest- 

 ing to botanists, I shall state that in so doing I have only 

 followed the example of M. Aug. de St. Hilaire, who, in his 

 Morphologic (p. 751), very positively noticed the relations 

 between the caruncle of Euphorbias and those which render 

 the seeds of Polygala so remarkable. 



What at first seems very curious in the seeds of the latter 

 genus, is the various forms and sizes of the caruncles in 

 different species. Always next the hilum, but independent 

 of the funicle, they are sometimes nothing but simple conical 

 tubercles, tridentate or trifid at their base ; one or two linear 

 fleshy extensions sometimes proceed from this same base, 

 and extend to a greater or less degree towards the chalaza, 

 being applied to the surface or back of the seed. But, not- 

 withstanding these variations in form, the caruncle is always 

 found at that point of the seed to which the radicle corre- 

 sponds, and the nearly constant direction of the latter towards 

 the micropyle is sufficient to make us suppose beforehand 

 that the opening of the integument is found on the excrescence 

 itself, or rather that the latter is nothing but the dilated 

 exostome. This also was observed by Aug. de St. Hilaire, 

 and if the opinion of so profound a philosopher were not 

 a sufficient guarantee for the exactness of the fact, I should 

 state that I have clearly seen the micropyle at the anterior 

 part of the caruncle of Polygala myrtifolia and speciosa. It 

 is then, I think, quite clear that the caruncle of Polygala is 

 completely analogous to that of Euphorbia, and this new 

 point of resemblance between the seeds otherwise so similar 

 of these two genera, explains better why Adanson brought 

 them both together in his family of Tithymales. 



In other Milkworts the caruncle is singularly modified ; 

 that of Comesperma is covered with long hairs which conceal 

 the whole seed ; the thick oily one of Badiera occupies the 



