iJ-'SO Mr. Tovey's Uese arches in the 



interims), or run in grooves (as the long head of the biceps), 

 or perforate other tendons (as the deep flexor of the fingers), 

 or turn through fibrous pulleys (as the digastric, the extensor 

 of the toes, &c.). By comparing the effect of a known force 

 acting on particular tendons, at first in their natural situations, 

 and afterward detached and free, the influence of friction in 

 each case would be readily determined. This source of error 

 seems to have been very generally overlooked by writers on 

 animal mechanics. 



I conclude, for the present, with suggesting that to distin- 

 guish the pectoralis major into "portio elevans " or " attol- 

 lens," and " portio deprimens," might serve to impress the 

 rationale of its peculiar insertion and twofold action, upon the 

 memory of the student. 



LXXIX. Researches iti the Undulaiory Theory of Lights in 

 continuation of former Papers. By John Tovey, Esq. 



To the Editors of the Philosophical Magazine and Journal. 

 Gentlemen, 



pXAVING deduced, (p. 500 of your last volume,) by a new 

 ^ method, the laws of the propagation of plane and sphe- 

 rical waves in elastic media, I will now, with your permission, 

 show how the formulae may be extended to the most simple 

 cases which are known to occur in the undulatory theory of 

 light, of waves not spherical emanating from a center of agi- 

 tation. 



(1.) It will be remembered that in my paper at p. 270 of the 

 last volume, the sums X were considered as comprised in three 

 classes, when it appeared that those of the first class, com- 

 posed of odd products of the differences, vanish, in conse- 

 quence of the first supposition there made respecting the 

 arrangement of the molecules. The sums also of the second 

 class, composed of even products involving odd powers of the 

 differences, were neglected ; because the terras of these sums 

 must be about half of them positive and half negative, and 

 consequently the sums themselves very small in comparison 

 with those of the third class, which last, being composed of 

 even powers of the differences, have their terms all positive. 



(2.) If the radius of the sphere of influence be not very 

 much greater than the intervals between the molecules, the 

 sums may or may not be sensibly the same for different direc- 

 tions of the coordinates, according as the intervals are the 

 same or different for the different directions. Suppose, for 

 example, every eight adjacent molecules to be at the corners 

 of a rectangular parallelepiped ; suppose fig. 1 to be a sec- 

 tion of the medium, the dots denoting the molecules in their 



