12 The Graspedosomaticlse of North America. 



through it or recover his powers at a temperature considerably 

 below the freezing point. This can be understood if we may sup- 

 pose that these animals have means of obtaining and using 

 larger amounts of ox3"gen than other tracheata, and are more 

 warm-blooded. In accord with this supposition is the following 

 observation on Trichopetalum by Dr. John A. R3^der, several 

 times verified b}- ourselves. 



" While examining some living specimens of Trichopetalum, 

 lunatum under the microscope, about a year since, I observed 

 that the respiration of the animal appeared to be conducted in a 

 most singular manner. The air seemed to be drawn in under the 

 labrum, and in some way to enter the dorsal cardiac sinus as 

 bubbles, which could be traced for some distance, more than half 

 the length of the animal, as they traveled slowly backwards, until 

 they disappeared over the opaque mass of ingested food contained 

 in the intestine. These bubbles of air always passed backwards. 

 It may be that they passed backwards inside of the intestine, but 

 the impression that I got was that they were traveling through 

 the cardiac sinus or dorsal heart of the animal. This circum- 

 stance may explain why it is that there are no pores on the side 

 of the body, though it is true that the lateral pores af millipeds 

 are usually ' foramina repugnatoria,' and have nothing to do with 

 the trachea or respiratory apparatus. This raises the question 

 whether our Lj^siopetalid^e are not distinguished from other 

 forms in some more important way than has been hitherto sup- 

 posed." 



In connection with this ability to withstand cold should be 

 mentioned the fact that in Europe the Alpine species affecting 

 the highest altitudes are Craspedosomatidse. 



The eyes are well developed in Trichopetalum, though the ocelli 

 are only moderately numerous. That is, the individual ocelli are 

 large, prominent and convex, as is usual in the family, which has 

 eyes apparentl}^ better developed than in other Diplopoda, and 

 not flat or merely pigmented spots as in other cases. But the 

 type of eye seems to be the same as in the lulidse and other 

 families, and the animals take no notice of anything in front 

 until the antennae are in contact with the object, indicating that 

 in all probability, the eyes are unable to form definite images, 

 but are merely' spots sensitive to light. Yet the animals do not 

 appear to have as decided objection to light as members of other 



