308 Mr. R. I. Pocock on Liphistius and its 



value of these peculiarities. Therefore the structural interval 

 between Li])Mstius and the Theraphosidfe, which have been 

 looked upon as its nearest allies, is greater than the interval 

 between the Theraphosidse and the Epeiridge, two families 

 which, omitting Liphistius^ lie at opposite poles of the order 

 Aranese. For striking and important as are the differences 

 between Theraphosa and Epeiraj so many intermediate genera 

 are known that it is almost impossible to give any one cha- 

 racter that will serve infallibly to distinguish the two sub- 

 orders of Araneai of which these two genera are types. 



The isolated position that Liphistius occupies with respect 

 to other spiders can perhaps be best expressed by setting it 

 apart by itself in a group equal in value to a group containing 

 all the others. For these I propose the names Mesothelee 

 and OpisthothelEe, the terms being derived from the position 

 of the spinning- organs. 



This removal of Liphistius from the vicinity of the Terri- 

 telarise is further supported by the fact that it shows more 

 than one hitherto, I believe, unnoticed point of resemblance 

 to the Dipneumonous spiders. One of these points is the 

 direction of closure of the mandibular fang ; the other, which 

 will require some elucidation, is to be found in the structure 

 of the spinning-mamniillEe. 



In Liphistius it will be remembered there are four large 

 and four small mammillae, the smaller being placed in pairs 

 between and a little in front of the larger. In the Dipneu- 

 mones there are two large and one small pair of mammillge, 

 the small pair being placed between and a little in front of 

 the mammillas of the posterior large pair. There can be no 

 doubt that the larger pairs of mammillas are strictly homolo- 

 gous in the two types just considered. Moreover I can see 

 no reasonable grounds for doubting that the intermediate pair 

 of the Dipneumones is also homologous to the posterior 

 auxiliary pair of Liphistius. This leaves the anterior 

 auxiliary pair of the latter animal to be accounted for. Now 

 in a few families of Dipneumones there is an additional 

 spinning-organ situated in front of the anterior mammilla 

 and known as the crihellum. This usually has the form of a 

 transversely elongated plate ; but in Filistata it might be 

 described as a large tubercle placed between the anterior 

 mammillse, the summit of which is divided by a longitudinal 

 groove into a right and left half (fig. 2). This cribellum, I 

 believe, is the homologue of the anterior auxiliary mamraillaj 

 of Liphistius joined together in the middle line. The double 

 origin of the plate is shown by a groove that marks the 

 surface upon which the spinuing-tubuies are situated. It 



