CrKKKNI LlTKRATURE. 259 



arouses some curiosity as to the considerations which lead the 

 pi blisher, John Murray of London, to send it to press. 



Prof. Ewart has lately published a study on the longevity 

 of seeds (Proceedings of the Royal Society, Victoria), in 

 which he corrects many of the earlier incorrect, contradic- 

 tory or misleading observations on the subject. He made 

 nearly 3000 tests and gives in addition the observations of 

 many previous workers. He finds that the best external 

 physical conditions for longe^•ity of seeds are a cool tempera- 

 ture, and dry, airy storage conditions. In the case of ordinary 



seeds the drier they are the longer they will last. The longest 

 lived seeds are usually provided with cuticularized, or more 



or less impermeable cuats. Howe\er, longevity does not 



depend on food material or seed coats, but is an hereditary 



peculiarity inherent in the protoplasm of certain seeds. The 



length of time that seeds will keep well varies from 36 hours 



drying outside the fruit {Cacao), to 1^0-250 years, which 



is the extreme for any known seed. 



Seeds which are adapted to longevity show no special 



adaptation to dispersal. They are, in fact, distributed in 



time rather than in space. The greatest number of long lived 



seeds are found among the Leguminosae. 



G. von Ihering, after prolonged observation in the \icin- 

 ity of Rio Janeiro, comes to the conclusion that the species 

 of Cecropia there cultivated have "no more need of protect- 

 ing ants than a dog has of fleas." What has so long been 

 held as a remarkable form of symbiosis he considers to be 

 simply a case of parasitism, since the death of a Cecropia 

 inbabitef- by ants Is followed by the death of the colony. A 

 young Cecropia which he planted and watched for H\e years 

 grew into a fine tree and remained entirely free from pro- 

 tecting species of ants. His observations, moreover, go to 

 show that the leaf-cutting ants are not destructive to vegeta- 

 tion, as has been supposed, and that leaf-cutting goes on in 



