24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 8o 



claw has become transformed into a featherlike empodium. In Japyx 

 it is reduced almost to a vestige (pi. 7, fig. 22), while in some Cain- 

 podea no vestige remains (pi. 7, fig. 22). 



A peculiarity in the two-clawed Collembola is that the smaller 

 element, which may or may not be a true claw, occupies a ventral 

 and a reversed position to the large claw. This smaller element is 

 referred to as the unguiculus by Folsom (1916), as the inferior claw 

 by Guthrie (1903) and as the empodial appendage by Carpenter 

 (1916). De Meijere (1901) has discussed its homology and calls 

 attention to the fact that it is not entirely ventral median in position. 

 Hansen ( 1893) considers the smaller claw a true claw and a primitive 

 condition from which type the two-clawed pretarsus of higher in- 

 sects was developed by a shifting in position and enlargement of this 

 element. By comparison with the claws of Symphyla and Pauropoda 

 one would be inclined to consider the inferior pretarsal element in 

 Collembola as one of the persisting lateral claws of the three-clawed 

 type of tarsus. In Symphyla the smaller lateral claw stands at about 

 right angles to the large claw. But in Collembola the inferior ap- 

 pendage is usually reversed and in some instances is detached from 

 the chitinous base of the large claw which alone usually bears the 

 tendon of the flexor muscles. Furthermore there exists a type of pre- 

 tarsus in Tomocerus (pi. 7, fig. 23) in which the large dorsal claw is 

 accompanied by two lateral claws springing from its base, in addi- 

 tion to the ventral appendage. These facts would indicate that the 

 so-called unguiculus, or empodial appendage, is not the true claw 

 but a secondary development, possibly only a modified seta. 



In the Crustacea, as stated by De Meijere, several genera of isopods 

 show the presence of a large and a small tarsal claw, as is commonly 

 found in Symphyla and Collembola. A drawing is here given of a 

 sow-bug tarsus showing the second claw (pi. 7, fig. 24). That the 

 two-clawed condition in the Crustacea should develop among land 

 forms is significant in considering the evolution of terrestial arthro- 

 pods. Probably it is correlated with the habit of crawling about 

 objects in a nearly vertical or even in an inverted position. The 

 almost universal ])resence of two tarsal claws among insects speaks 

 much in favor of this arrangement for an efficient clinging organ. 



POSTCKPHALIC BODY-SEGMENTATION IN ARTHROPODA 



Theoretically arthropods, as we know them today, have been con- 

 sidered as descendants from a type in which the whole body consisted 

 of a series of similar segments arranged end to end along the body 



