AMERICAN COLEOPTERA. 91 



their number the last ventral segment is usually said to be " sis- 

 spinous." The last dorsal however undergoes the most important 

 changes, all however traceable to the development of a primary 

 simpler form. 



The simplest type of structure seems to be that of marnlicoUiii, in 

 which the last dorsal is deeply divided at middle forming two tri- 

 angular processes broad at base, acute at tip. The first modification 

 of this appears in limbatus, iu which these processes become more 

 slender and between them appears an oval plate acute at tip and about 

 half their length. This middle lobe next becomes elongated so as to 

 equal the lateral lobes and is bifid for a short distance at tip {frigidus), 

 or becomes cleft entirely to base forming slender processes, similar to 

 those at the sides. These in turn become one-fourth shorter than the 

 lateral processes although slender, as in migustatus and several others. 

 As these processes become still shorter they also become broader and 

 finally unite forming a trapezoidal plate acutely notched at tip as in 

 fimhriatua. The next change is in memnuvius in which the plate 

 becomes narrower, more elongated and with a small notch at tip, and 

 finally it becomes as long and nearly as slender as the lateral teeth 

 with the tip acutely divided as in caiutdensis and tacltijporokles the 

 division being at times (iiisfahilis), so minute that the middle process 

 is slender at tip, similar to those at the sides and equal in length. 

 The middle process while remaining slender has a lateral tooth arise 

 on each side, acute and much shorter than the main stem, as is seen 

 in Crofchii The three processes now (instabilis), approximate and 

 finally {nii/ricornis, luritfus, circumcinctuii) unite, being separated at 

 tip by a slight notch and a groove on the upper surface, the middle 

 lobe is prominent and acute in riigricornis and (uridus, or broader and 

 emarginate at tip in circumcinctuii. 



These sexual modifications of Taciiinus form a very interesting 

 study, without a knowledge of them it is impossible to separate the 

 species correctly. 



The following tables are based on the characters above mentioned, 

 the males and females being considered entirely apart and independ- 

 ently of each other, although it frequently happens that the one sex 

 of two species may be very similar while the opposite sex is altogether 

 different. 



