356 NATURAL SCIENCE. Nov.. 



the metacarpal bones support i — 6 (metacarpals) the single phalanx of 

 digit III. bears no. 7 (addigital) ; 8, 9 are attached to phalanx I. of 

 digit II. (middigitals),and the remaining feathers to the antepenultimate 

 phalanx of digit II. (predigital). This is the normal arrangement as 

 we find it to-day. Bands of muscle and tendon bind these feathers 

 together, and to the skeleton, so firmly that a large amount of force is 

 necessary to remove any one of them. The whole of this knitting 

 together is completed by a strong tendon running along the posterior 

 (postaxial) margin of the wing some distance down the quill, from the 

 elbow to the tip of the longest finger, the quills appearing much as 

 though they had been pushed through it, so as to ensure their 

 preserving the right distance apart, and all opening and closing in 

 unison. The quills of the hand are set rather closely together, so 

 much so, that they nearly touch one another at the base, but not 

 quite ; thus a series of troughs is formed, and in these lie the bases of 

 the strong, stiff, major coverts which therefore serve as an additional 

 source of strength, binding all into a compact whole. 



Turning to the restoration, it will be seen that I have assumed 

 some such arrangement to have obtained in Archaeopteryx, as that 

 which has just been described. Indeed, the number and position of 

 the digits, the proportion, texture, and position of the feathers, are so 

 clearly indicated on the slab, that I should rather be said to have 

 copied than " restored " this wing, using that of living birds as a guide 

 to the details. But let the reader turn to the photograph (PI. I.) and 

 judge for himself. First of all, I would suggest that he should begin by 

 " setting " digit III., which has evidently been dislocated. Then, if 

 the line of the shafts of the quills be carried up toward the skeleton, that 

 of the outermost and shortest will be found to lead to the tip of digit 

 III. running alongside the claw; tracing up the others in like manner, 

 he will find that he has room proximally for some two more feathers, 

 which I have taken the liberty of adding ; the fact that these are 

 not indicated on the slab is probably due to the cubital remiges over- 

 lapping them. 



We have now to consider the question as to whether the actual 

 feathers (or their equivalents from infiltration) are preserved on the 

 slab, or whether what is left is but the impression of the ventral 

 surface ; either of these two views is possible. I do not remember to 

 have seen this point raised before, but the general impression among 

 my friends seems to be that, what we have presented to us is the 

 ventral surface. If this is the case, as my colleague Mr. E. S. Good- 

 rich has pointed out to me, not feathers but only their impression is 

 left on the slab. 



I should be delighted if it could be proved beyond doubt that 

 this latter view is correct, for we should be more easily able to under- 

 stand the long, semiplumous covert feathers partly concealing the 

 remiges, inasmuch as the coverts of the under surface are often of pre- 

 cisely this texture in modern birds. Again, this would account for the 



