ON THE CHILOPODA OF NORTH AMERICA. 17 



Fam. IV. SCOLOPENDRIDiE, Leach.* 



Segments ppdophora 21—23. Oculi pauces vel nulli. Pedes postretui incrassatre plerumque spinosi. 

 In this family the principal generic characters are founded upon the number of 

 segments of the body, the shape of the head, the number and form of the spiracles, 

 the absence or presence of eyes, and the form of the terminal segment and 

 its appendages. In some cases the number of joints of the antenna} seems to be a 

 constant generic subcharacter. In the large genus Scolopendra— the Titans of the 

 Myriapoda — the principal specific characters are founded upon, first, the number of 

 segments to the antennas; secondly, the number and arrangement of the labial teeth ; 

 thirdly, the peculiarities of the posterior feet ; fourthly, the shape and comparative 

 size of the head. We have no doubt that the number of joints to the antenna) is 

 fixed for most species, but it serves more generally to distinguish groups than 

 single species; yet it occasionally is the most reliable character separating closely 

 allied forms. Unfortunately it must be used with great caution in the identification 

 of individuals, for, owing to the ease with which portions of the antenna) are lost, 

 the want of a certain number is a very indefinite negative character. After de- 

 taching a few of the distal joints, no trace of their former presence is left. For 

 the same reason much caution is also necessary in assigning the number in a de- 

 scription. M. Saussure has found so much variation in this character that he considers 

 it worthless. It is possible that it may vary in certain species ; but the differences 

 generally consist in there being too few joints, which, as we have shown above, is to 

 be looked for, and reproduced antennas probably have occasionally an abnormally 

 great number of very small ones. The number and arrangement of the labial teeth 

 are a good character, but there is often an agreement between distinct forms, and on 

 examining large suites of individuals, we have found more variance in tiie same 

 species than our reading would lead us to look for. The posterior legs furnish the 

 best criteria in the distinguishing of species. Most species have peculiarities either 

 in the shape or relative size of the joints, or in the number or arrangement of spines 

 on them. In order to show the amount of constancy that these characters possess, 

 we have drawn up a series of tables, which may aid in establishing their value. The 

 color, we think, is not to be depended upon, although Mr. Newport seems to attach 

 some importance to it. In the preservation of specimens it is very liable to be altered 

 or destroyed ; besides this fact, our studies of these animals, both in museums and, to 

 a limited extent, in the Tropics, lead us to the opinion that the color varies exceed- 

 ingly, even during life. Size is often a good distinguishing character of a species, 

 although scarcely available for the identification of an individual. 



If we consider a species as the expression of a preconceived idea, there must be, as 

 it were, a type of every species around which the individuals cluster, restrained from 

 departing beyond a certain limit from the central nucleus. We can see then how 



* Linn. Trans. xi. 

 5 



