hearing upon the Classification of Spiders. 313 



dants of the Pedipalpi and the latter of the Scorpions ; or, to 

 put it differently, that of existing Arachnida the Pcdipalpi 

 come nearest to the immediate ancestors of s])iders and the 

 Scorpions nearest the ancestors of the Pedijialpi. I hope in 

 a subsequent paper to work out the classification of Arachnida 

 from this standjioint. At present it will be sufficient to state 

 that the primitive nature of the structure of Scorpions is 

 shown by the metameri-im of the body, tlie serial repetition of 

 similar somites being carried to a greater extreme than in 

 any other order of Arachnida. 



This then being my belief as to the ancestry of the 

 Araneaj, I see no escape from the conclusion that Upliistiiis 

 is a transitional form — a missing link — between the Opistho- 

 thelse and the Phrynidte. Certain it is that Liphistius 

 possesses at least two important permanent characters which 

 are only found in the embryos of other spiders. These 

 characters are the segmentation of the abdomen and tlie ante- 

 rior position of the external spinning-organs. As is well 

 known, these organs are the third and fourth pairs of abdo- 

 minal appendages, which are primitively situated in a line 

 with the first and second pairs on the lower surface of the 

 anterior half of the abdomen. The migration of these appen- 

 dages to the posterior end of the body, which takes place in 

 all spiders except Li])/tistiiis, is a secondary modification 

 which is no doubt beneficial as conferring a greater freedom 

 and range of movement upon organs requiring considerable 

 manipulation. 



Liphistius, then, retains certain embryonic characters that 

 all other spiders lose ; we may conclude therefore that the latter 

 are " higher " than the former. Of the other spiders, those 

 that on the whole come nearest to Liphistius are the Mygalo- 

 morpha3. These therefore are " lower " than the Arachno- 

 mor))hie ; and the lowest of the Arachnoraorphse are Jlypo- 

 chilys, J Jy sclera, and Filistata. 



Since, theti, some reasons have been shown for thinking 

 that Liphistius is of living spiders the nearest to the ancestral 

 form, and, secondly, that this spider possesses the homologue 

 of the cribellum, we can without difficulty explain the exist- 

 ence of this organ in widely different genera, and its presence 

 at once loses the systematic importance that Dr. Bertkau 

 and Mons. Simon have claimed for it. 



The same argument will apply to the presence of two or 

 three claws on the feet of the OpisthothelEe ; for since Liphis- 

 tius possesses three well-developed claws, the third claw may 

 have been retained or lost indiscriminately, so to speak, in 

 different genera. So that Bertkau's subdivisions of Ecri- 



