THE FLORIDA STRAXGLING FIGS. 29 



It is a fact that the long-stalked; short-styled flowers are pre- 

 ferred by the ovipositing Blastophaga, for only rarely can one 

 find more than one or two such flowers producing seeds, they 

 being usually occupied by Blastophaga larvae (plate 8, fig. 7), 

 However the fact that such flowers of this type as escape ovi- 

 position do develop perfect seeds demonstrates that they are 

 not speciaUzedgall flowers. Furthermore, of the short-stalked, 

 long-styled flowers always a certain proportion, varying from 

 5 to 20 per cent, j conisUn Blastophaga larvae, yet there are no 

 observable differences in this type of flowers at the time of 

 pollination. 



The single stamen is plainly divided into filament and 

 anther in Ficus populnea (plate 8, fig. 10), and is larger than 

 that of F. aurea in which the anther cells lie upon the top of 

 a somewhat obliquely truncate column (plate 8, figs. 4-G). 

 The calyx lobes in both staminate and pistillate flowers arc 

 smaller proportionally and more distinct in F, aurea. They 

 are mostly three in number in both kinds of flowers. In F. 

 populnea, at least in the specimens studied, the calyx lobes 

 seem to be two in number in the pistillate flowers, and come 

 well up above the top of the ovary. Possibly the larger 



lobe 



I 



flowers but one large lobe could be found. Whether it 



■ 



represents a union of several lobes or not was not deter- 

 mined. In both species the ripe achenes have an outer layer 

 of cells that form mucilage in water. On drying, this serves 

 to glue the achenes in place, 



A study of the germination of the seeds of Ficus aurea 

 revealed the very interesting fact that it is dependent upon 

 light, not taking place if the achenes are left dark. The 

 following experiments were performed in this connection : 



October 22, 1907, several ripe fruits were opened and the 

 achenes and surrounding pulp together with empty galls 

 were placed in two Petri dishes upon filter paper moistened 

 with Knop's mineral nutrient solution. One dish was placed 

 on a table next to a north window, the other in the back of a 

 long drawer in the same table. In the former the first signs 

 of germination were observed November 1 and by November 

 6 apparently all the seeds had germinated. In the dark there 



