1890.] OF THE FAMIL\ BUTHID^. 115 



indeed, there are some grounds for thinking that this may have taken 

 place in the ease of Grosphus and of Rhoptrunis ; for the former 

 appears to connect Leprous with Buthus, and the latter Isometrus 

 W\t\\Buthns ; or, in other words, Buthus appears to have arisen from 

 Lepreus by way of Grosphus and from Isometrus by way of Ehop- 

 trurus, that is from two independent sources. And if anyone likes 

 to believe that this has t^ken place, it is difficult to see how the idea 

 can be shown to be wrong. Of course an alternative hy])othesis, 

 namely, that Grosphus is the ancestor of both Buthus and Lepreus, 

 at once suggests itself ; but in that case it is hard to see why Lepreus 

 should have lost the two mandibular teetii. which must surely be of 

 considerable service in the battle for life. Moreover, when we reflect 

 that Lepreus agrees with almost all the Scorpionidoe (including pro- 

 visionallv Vejovis and Bothriuriis) in the absence of tliese teeth, it is 

 hard to believe that it is not a character which has been transmitted 

 to Lepreus from some unknown member of this family. In that case 

 we must, it seems to me, account for the resemblance between Grosphus 

 and Lepreus on the hypothesis that the latter is the ancestor of the 

 former, unless, indeed', we consider that it is the result of what, for 

 want of a better term, may be called accident. However, from 

 whichever side the question be approached, some obstacle presents 

 itself which our knowledge of the affinities of the genera is at present 

 too limited to surmount." For a variety of reasons, however, it seems 

 to me to be perhaps well to regard provisionally Lepreus and Uro- 

 plectes as derived from Grosphus; for undonbtedl}' in most respects 

 these two genera dej)art widely from a plan which is common to all 

 the others. With the exception of these two and of Butheolus, a 

 genus hard to locate, the accompanying pedigree (see p. 128) appears 

 to me to represent fairly well the mutual relationship of the genera 

 and subgenera here recognized. But it must be regarded as merely 

 tentative and in no way as expressing a final opinion. 



Considering the Scorpionidse as a whole and the Buthidse as a 

 whole, and noting what characters are common to both and what are 

 the average cha~racters of the least specialized of the genera of 

 Buthidae, we are able to form some opinion as to the characters of 

 the immediate ancestor of the Buthidae, or, in other words, to discover 

 the common plan from which all the modifications of the various 

 genera can be derived. 



By this means it may be inferred that in this hypothetical ances- 

 tral form the sternum was triangular ; the movable digit of the cheli- 

 cerfe was furnished with three teetli above and two below (not counting 

 the terminal fang), the immovable with a single row of teeth ; the^ 

 arniature of the digits of the chelce was composed of a number of 

 oblique, parallel, slightly overlapping rows of denticles ; there were two 

 median eyes, and threelateral eyes on each side ; the cephalothorax 

 was granular, but not carinate ; the tergites were granular and fur- 

 nished with a median keel, the last, in addition, bearing two lateral 

 keels ; the sternites were smooth and anteriorly bisulcate, the last 

 only being furnished with two or four keels ; the tail was keeled 

 throughout, and there was probably a spine beneath the aculeus ; 



