91 



buttress rises to the apex of the ridge. The crest is hollowed 

 above from before backwards, leaving the margins rather 

 sharp, or forming, as above described, a double ridge. These 

 edges rise into a series of teeth, a small and low one near the 

 palpi, then four of about equal prominence, but each higher 

 above the head than the previous one, so that the last is a pro- 

 minent spine over the vertex. The ridge then falls rather 

 abruptly and terminates with a linal spine close to the back of 

 the head against the prothorax. The corresponding spines on 

 the two margins are connected across the intervening groove 

 by ribs that break the groove up into a series of hollows or 

 basins. 



"The other portion of the apparatus is the first tibia. This 

 is short — ^just half the length of the second and third tibia. 

 The usual tibial spur arises at about one third of its length 

 from its base, and is fairly long (half length of tibia), but it 

 is curiously pale and flimsy-looking, and it is doubtless quite 

 functionless as far as regards the special use for which the 

 tibia itself is modified. Beyond the origin of the tarsus, the 

 tibia is prolonged into a long (half length of tibia) and a 

 short spine, with a sharp knife-edge running from the one to 

 the other. The long one is anterior and very sharp. The 

 tarsus flexes in the tibia readily, but admits of very little 

 extension, thus avoiding getting in the way in the special 

 use for which the tibia is designed. The femur is very thick 

 and short, obviously strong and muscular for the exertion of 

 the necessary power in cutting the cocoon. 



" I have not seen the apparatus at work, but taking its 

 structure in connection with the form of the slit in the 

 cocoon, it is obvious that the cutting is done chiefly by the 

 tibia, both apparently acting together, as the slit is single, 

 except that it sometimes divides into two branches at its 

 lower end ; the long sharp spine penetrates the cocoon and 

 the short one prevents the tibia passing too far through, and 

 the cocoon is then cut against the short ridge joining the 

 spines as the tibia is forcibly pressed downwards. I ought 

 to have said for clearness that these two spines are prolonga- 

 tions of the tibia itself, and not jointed appendages, as one 

 usually understands by tibial spines or spurs. The even 

 more remarkable armature of the head presents a series of 

 points by which the cocoon is firmly held and steadied 

 against the strain of the tibial pressure. The tibiae are thus 

 able to cut the cocoon instead of pressing it before them. 

 These spines also probably do some cutting. It is worth 

 noting that the most dorsal of these spines is so placed that, 



