Structural Analogy. 359 



thereby enabled to tread securely upon lightly compacted 

 snow. 



The problem of walking safely upon the yielding surface 

 of snow has been solved by grouse as by man — by a large 

 increase in the supporting area brought into contact with it. 



Among that interesting group of marine animals known as 

 the Crinoidea a very similar problem has been solved in a 

 strikingly similar way. The Crinoidea are dependent for 

 their food upon small organisms which fall upon their up- 

 turned and outstretched arms and pinnules, and are thence 

 conducted by means of a ciliated " ambulacral groove " to the 

 mouth. It is at once evident that if the lateral area of the 

 very slender pinnules can be increased in any way, the 

 number of falling food-particles intercepted, and hence the 

 available food-supply of the animals, will be proportionately 

 increased — the same question arising in regard to the presenta- 

 tion of a maximum food-collecting surface directed toward the 

 rain of food from above among the crinoids as arose among 

 the grouse in reference to the development of a maximum 

 area of support upon a loosely compacted surface. In the 

 crinoids, as in the grouse, two rows of supplementary plates 

 on either side of the single median row have been developed, 

 the first row of squarish plates, the outer of elongate rounded 

 plates, extensible laterally, but capable of being folded in- 

 ward over the delicate ambulacral groove, so as to serve as a 

 protection to it, an additional service not required in the outer 

 series of scutes in the grouse. 



I have been unable to find in the literature any account of 

 the origin of the scutes in the grouse ; but Dr. Stejneger has 

 suggested to me that they are in all probability merely modi- 

 fied feathers, a derivative from some such condition as that 

 found in the Red Grouse (Lagopus scoticus) or Ptarmio-an 

 (Lagopus mutus). In the crinoids these two rows of supple- 

 mentary plates are formed directly from the pinnule segments ; 

 the distal angles of these expand and project in the form of 

 thin rounded flange-like processes, which eventually become 

 separated off from the pinnule segments by suture. In some 

 groups there is no further development; but in others the 

 distal outer angles of the plates of these newly formed rows 

 ("side-plates") become expanded, and these again become 

 separated off from the parent plates by suture, forming the 

 second row of laterally extensible rounded plates (" coverin ex- 

 piates"). 



Now it has been found that, as a rule, these side- and 

 covering-plates are much better developed and of far more 



