606 



into three, as he considers Philodromus and Stephanopis as types of 

 separate families. It now appears to me also reasonable to divide 

 this sub-order into four families; but I prefer to take Heteropoda , 

 instead of Philodromus, as the type of a new family, and I accor- 

 dingly propose to divide the Laterigradce into Thomisoidce. , Hete- 

 ropodoidce, Staph anopoidce and Arehnoidce , and the Thomisoidce 

 further into Thomisince, Philodromince and (?) Anetince. The genera 

 Philodromus , Artanes and Thanatus seem to me to be in fact more 

 nearly allied to the typical Thomisinse than to Heteropoda, Microm- 

 mata, Sparassus, Selenops, Delena, Voconia, JBemiclcea etc., which I 

 combine under the denomination of Beteropodoidce. The distinguish- 

 ing characteristics of this group may be easily gathered from what 

 I, On Bur. Spid., pp. 173 et seq., have advanced of some of the 

 genera belonging to it. As regards Stephanopis, vid. Cambridge, Descr. 

 and sketches of some new species of Araneidse, with characters of a 

 new genus, in Ann. and Mag. of Nat. Hist., 4 Ser., Ill, p. 60 (9); 

 Thorell, Aranese noimullse Novae Hollandise, p. 378. 



Of the sub-order Citigradon no remarkable forms have, as far as 

 I am aware, been of late years discovered, neither have any altera- 

 tions of any importance in the classification of this sub- order been 

 proposed. (See however above, p. 598). 



In the sub-order of the Saltigradce a separate family, Aphanto- 

 chiloidce, ought perhaps to be formed for the genus Aphantochilus 

 Cambr. ') (referred by Cambridge to the Myrmecides) , which differs 

 from all hitherto known spiders by the absence of the labium. — 

 Cambridge 2 ) thinks, that the family OtiothopoidcB Thor. ought to be 

 suppressed and Otiothops referred to the same family as Palpimanus, 

 because the abnormal number of joints in the legs — on which I 

 laid especial stress — cannot be a matter of sufficient weight to form 

 the foundation, on which to erect a family, but at the most a genus. 

 There is however a considerable difference between one of the tarsal 

 joints being divided into two, as may be the case within the family 

 Hersilioidce, — which Cambridge adduces as an example of a family, 

 in which genera with different numbers of joints in the legs occur, — 

 and another of the joints of the legs being entirely absent, as is the case 

 in Otiothops. The latter circumstance appears to me of far greater 

 systematic importance than a deviation from the usual number of 



1) On some new gen. and spec, of Aran., p. 744. 



2) Bibliogr. Notice etc., p. 417. 



