EVOLUTION 971 



of three links between drum and hearing ear can be traced back to 

 the sturdy hinge-work of the jaw in fishes. It should be kept in 

 mind that the lower jaw of mammals has become a single bone on 

 each side — evolution by simplification — and that it articulates 

 directly on the skull with the squamosal, not with the intermediary 

 quadrate, as in Birds and Reptiles. This change has set a number 

 of hinge-bones free ; and the interesting point is that in their func- 

 tion-change they have become very small and yet highly specialised. 

 They have been utilised for a more delicate function, and this is 

 reflected in the nicety of their structure. Here, again, we find an 

 illustration of what seems to have been a frequent method in 

 evolutionary change. Structures fall into desuetude and dwindle; 

 but instead of becoming vestigial, they are often reinstated in 

 service with new functions. 



(5) On a different line is the origin of a novel structure by the 

 exaggeration of part of an old one. The trunk of the elephant, with 

 its extraordinary muscularity and tactility, is one of the clearest of 

 instances ; for what is it but a great exaggeration of the nose and of 

 a portion of the upper lip ? We see hints of the beginning of a similar 

 proboscis in the tapir and in the little aquatic insectivore, the 

 desman, which used to occur in England. Moreover, from the un- 

 usually complete series of forms ancestral to the modem elephant, 

 we know that the trunk, as testified to by features in the fossilised 

 skulls, was the outcome of a gradual change extending over millions 

 of years. 



(6) The spinnerets of spiders are transformations of the two last 

 pairs of appendages on the abdomen of the embryo, while the four 

 pairs in front of them simply disappear. The strange comb-like 

 sensory structures projecting on the under side of the third segment 

 of the scorpion's abdomen are the transformed third pair of 

 abdominal appendages, and the second pair, much reduced, form 

 the genital operculum. 



The "poisers" (halteres) of two-winged flies or Diptera, easily 

 observed in daddy-long-legs and blue-bottle, are two rapidly 

 quivering sensory structures, somewhat like the head end of small 

 pins. They are present in almost all Diptera, though rarely present 

 in wingless forms; and they are not known outside the Diptera 

 except in the male Coccid insects. It would be diflicult to find more 

 characteristic structures, and they are very intricate. Four muscles 

 enter the base and keep up a very rapid vibration of the "halter". 

 They are rich in nerve-endings and they have a somewhat ear-like 

 (chordotonal) structure at their base. Their precise function is still 

 obscure, but they are undoubtedly sense-organs. Now there seems 

 no doubt from their position and development that they represent 

 transformations of the second pair of wings, so normal in other 

 orders of insects. Their place is never taken by wings in Diptera, 



