NO. II LEGS OF PRIMITIVE ARTHROPODS EWING 23 



it is vestigial and in some few clearly wanting. A complete series 

 exists among the various species showing the reduction from the 

 normal condition for the group to the smallest vestige and to a one- 

 clawed condition. 



The best evidence as to the nature of the primitive type of insect 

 claws is to be found, it is believed, by a comparative study of the 

 claws of primitive insects with those of the primitive arthropods most 

 closely related to them. When this is done we find in all instances, 

 the present writer believes, that either the three-clawed type exists, 

 as found in most Thysanurans, or a two-clawed or one-clawed type 

 the origin of which may be clearly traced from the three-clawed type. 



In pauropods the generalized legs all show a tarsus (pi. 7, fig. 20) 

 terminated by a large central claw and two small, lateral and fre- 

 quently unequal, claws which spring from a basal foot-pad at almost 

 right angles to the central claw. There is a tendency among the 

 generalized legs in Pauropiis for one of thd lateral claws to become 

 vestigial, and this tendency reaches a climax in the last pair of legs 

 of Pauropus (pi. 7, fig. 20, C) where one claw is entirely lost. 



In Symphyla the different genera show a marked similarity in re- 

 gard to the tarsal claws. The prevailing type is a tarsus with two 

 unequal claws (pi. 7, fig. 20, D), a type comparable with that of the 

 posterior legs of Pauropus. However, the tarsi of the first pair of legs 

 in some symphylids as well as those of some of the other legs show 

 three claws, one of the lateral ones being reduced and modified so as 

 to be somewhat setiform. 



In the Protura only a single claw has been described, but Berlese 

 (1910) pointed out that there exist rudiments of a second claw on 

 the first tarsus of Acerentomon doderoi. It is observed, further, that 

 in one of our American proturans, Acerentulus barberi, two func- 

 tional claws are present on tarsus I (pi. 7, fig. 23). In this species 

 the large claw is posterior in position, the small one is middle, and 

 a claw rudiment is anterior. This small claw is somewhat setiform, 

 but its position and articulation indicate its homology. In those tarsi 

 that have only the single claw, this claw usually occupies a posterior 

 position indicating its homology with the large claw of the first tarsus. 

 The dominance of the posterior of the three primitive claws in the 

 Protura appears to be quite a different condition from that which is 

 found in the Collembola. 



The two-clawed tarsi of Campodca and certain other thysanurans 

 are explained by Snodgrass (1927) as being a derivation from the 

 three-clawed Lepismid type. Here it appears evident that the middle 

 claw has disappeared. In some members of Campodca the middle 



