CALIFORNIA ACADEMY OF SCIENCES. 289 



tion of anchylosis; they are about equal in length and 

 together scarcely longer than the fifth. I then examined 

 the anterior tarsus in a series of specimens of Falorjria 

 keviuscula, Lee, in which the tarsi are very long and slender, 

 differing from the comparatively short and robust ones of 

 Colusa. To my surprise it could be plainly seen, when the 

 tarsus was held in a suitable position, that this also is five- 

 jointed, the fourth joint being very small and anchylosed 

 with the fifth; the suture is very distinct on the upper sur- 

 face, but is not so clear at the side, and if the tarsus were to 

 be detached, mounted in Canada balsam, and viewed later* 

 ally by transmitted light, it would probably be entirely ob- 

 literated. 



If the European species of Falagria have the anterior 

 tarsi in reality four-jointed, that is if the fourth from the 

 tibia has entirely disappeared by amalgamation, Ave should 

 have three typical genera, Falagria, Colusa and Autalia, ex- 

 hibiting so close a mutual resemblance as to indicate almost 

 unmistakably descent from a common ancestral type, and 

 having the tarsi 4-5-5, 5-5-5 and 4-4-5-jointed respectively; 

 in other words these genera would be representatives of the 

 three great groups of the tarsal nomenclature. 



In reasoning upon this we cannot but be persuaded that a 

 system of classification assigning these genera to groups in 

 which they are placed with others of very different lines of 

 development, is somewhat artificial and delusive. In this 

 special group the tarsi appear to be the organs which are 

 most susceptible to modification by descent, and therefore 

 should not be taken as a basis of systematic division. A 

 classification based upon the form of the head and its attach- 

 ment by a narrow neck or otherwise to the prothorax, would 

 give us two quite natural divisions of the Aleocharae, which 

 might then be subdivided according to modifications of the 

 antennae or of pronotal configuration, and, without claiming 

 that such a system would be better than that now in vogue, 

 it would certainly be more natural, and what is of no little 

 importance, it would be easily observable. 



