286 ./. K. ThacJier — Median and Paired Fins. 



First Dorsal of Mustelus cants, PI. XLIX and L, figs. 5-10. 



Id pi. XLIX, fig. 5, we have 24 separate rods, unless ?> be the terminal piece of 2 ; 

 but its conformation seems to testif}' to its independent but reduced character. The 

 number here then is 24 or 2,3. 



In PI. XLIX, fig. 6. we have again the same alternative, witliout quite so strong a 

 case for 24, but still quite strong. 



In PI. L, fig. 7, we have the choice between 24 and 2.5, but in favor of the latter. 



In PI. L, fig. 8, we have 23 or 24, but the former has the greater probability. 



PI. L, fig. 9 gives us 22, 23 or 24, 23 being more probable. 



PL L, fig. 10 exhibits 23 or 24, dependent on the view taken of rays 5, 6, and 7. 

 The probability seems in favor of 24. 



I think we may sum up then with regard to the number of rays constituting the 

 first dorsal of Mitslelus canis thus : it has usually 24 rays but this may vary to 23 or 25. 



Nearly all the rods are segmented twice. The distal line of segmentations fails in 

 the one or two orad; and the proximal, in the two or three aborad ones. Additional 

 segmentations are ver}' rare. What might be reckoned as such are seen in fig. 7, ray 

 4 ; fig. 6, rays 5 and 6. This makes an average of ■() + .* 



The union of adjacent rays is rare. I estimate it at '04 of the total possible con- 

 crescence. f 



Shortening or reduction in size is likewise rare. We have first those questionable 

 cases of which fig. 5, rod 3 seems the least questionable ; and then plainer but less 

 extreme instances in fig. 7, ray 5; fig. 9, ray 18; fig. 10. ray 7, then we have the usual 

 shortening of the rays at the extremities of the fin. When these aborad rays shorten 

 up, those next in front of them have a remarkable tendency to grow up under them. 

 This is well shown in fig. 9. 



iforeover when in the aborad rays the proximal joint becomes very short, it is some- 

 times divided into two lateral halves. This is the case for example in rod 22 of fig. 10. 



Occasionally we have a minute piece or pieces of cartilage forming a tip to a ray. 

 It cannot act as an extra joint, by giving increased flexibility to the ray. And it 

 seems doubtful whether the origin of the two is to be referred to the same causes. 

 Yet intermediate forms occur so as to raise the question whether they are to be 

 referred to one or the other category. Tliese tips seem to ))e exhibited in fig. 7, ray 4 ; 



fig. 8, rays 2 and 3 ; fig. 9, ray 3. This gives „ — - — + . 



I now find the ratio of the proximal piece of the middle ray of each tin to the 

 middle piece of the same. The average of these ratios is -6. The method gives -3 as 

 the ratio between the terminal and middle joints. 



We have then for the first dorsal of Mustelus canis : 



Number of rays 24. Extra segmentations '0 + . 



Concrescence '04. Betipping -0 + . Katio of proximal to middle piece of middle 

 ray -6. Ratio of distal to middle piece -3. 



* The decimal is obtained by dividing the number of additional segments by the 

 number of rays. 



\ The amount of concrescence between two adjacent rays is the ratio between the 

 length of the union and the whole distance through which they are adjacent and 

 might have united. The sum of these fractions divided by the number of rays less 

 the number of fins, gives the estimate of the concrescence. 



