302 The Botanical Gazette. [November, 



individual parts may be slender though the projectile is heavy, 

 since the strain exerted by the force is distributed through a 

 greater length, and is not so violent at a particular point. 

 But the strain is very great when a ball is struck by the end 

 of a bat, as would be the case with the rays of the umbellets, 

 for the center of percussion, where a blow is most effective 

 and where the strain is least, is considerably below the end 

 of a body of nearly uniform thickness. But observation 

 shows that the ends of the rays are not broken in the disen- 

 gagement of the fruit. 



Wh 



<T 



the Umbellifera is doubtful, because, if frequent, it probably 

 would have been observed in other cases. However, more ex- 

 tensive observations on this point are needed before express- 

 ing a decided opinion. But as yet I have seen none that acts 

 in the same manner. But the way in which the carpels split 

 off from the carpophore and are suspended from it must in 

 general facilitate the dispersion of the fruit of Umbelliferae 

 by shocks received from passing animals, and doubtless from 

 gusts of wind, though not acting like a sling as in Cryptota?- 

 nia. In an article by Hildebrand on this subject, mention is 

 made of a single genus of Umbellifers, Scandix, in which 

 the fruit is thrown from its support. But this is due to the 

 tension to which the dry fruit is subjected when adhering to 

 the carpophore, which causes it to be thrown to the right and 

 left through its elasticity when released from stress, and is 

 compared to that which occurs in Er odium. l This is the 

 only case among the Umbelliferae of fruits flying off by means 

 of their elasticity cited in his more general treatise, "Die 

 Verbreitungsmittel der Pflanzen," though several other modes 



of dispersion, with examples in illustration, are given for this 

 family. 2 



Englewood, Chicago. 



1 Die Schleuderfriichte und ihr im anatomischen Bau begrUndeter Mechan- 

 lsmus. (Jahrb. fur wiss. Botanik, ix. 270.) 



2 1. c. p. 140. 



