252 EXPERIMENTS ON MELONS, CYCAS, ETC. 



green and vigorous, nor did they begin to fade till 

 they had thus lasted for a very long while. Since 

 I read the history of this experiment, I have found 

 it easy in many plants to tell by the appearance 

 of the stigma whether the seed be fertilized or not. 

 The above experiment is the more important, as the 

 abbe Spallanzani has recorded one made by himself 

 upon the same species of plant, with a contrary 

 result. But as he has said nothing of the appearance 

 of the stigmas, his experiment must yield to that of 

 Linnasus in point of accuracy ; and even if his ac- 

 count be otherwise correct, the result is easily ex- 

 plained. Hemp, Spinach, some Nettles, &c., na- 

 turally dioecious, are occasionally not completely so, 

 a few latent barren or fertile flowers being frequently 

 found among those of the other sort, by which pro- 

 vision is made against accidents, and the perfecting of 

 a few seeds, at any rate, secured. 



In general, germens whose stigmas have not received 

 the pollen wither away without swelling at all ; but 

 some grow to a considerable size, and in such the sub- 

 stance of the seed, its skin, and even its cotyledons, 

 are often to be found, the embryo only being wanting. 

 In a Melon or Cucumber it is common to find, among 

 numerous perfect seeds, many mere unimpregnated 

 husks. In the magnificent Cycas revoluta which bore 

 fruit at the bishop of Winchester's, and of which a 

 history with plates is given in the sixth volume of the 

 Linnaean Society's Transactions, I found the di'upa 



