LE. Nipher—American Trotting Horse. 45 
. d 
Plotting the values of and the corresponding values of s, 
he then goes on to say, that the points so determined may be 
represented by a curve such that the value of aT increases 
when that of s diminishes. Assuming as I had done, that a 
straight line will represent the values, he determines the values 
of the constants, and finds that the line intersects the s axis at 
& point where s is —25. This would mean that the horse 
would finally trot a mile in less than no time. 
hen making the first discussion of the subject, the writer 
‘3 
considered the propriety of determining av by means of conse- 
values of - are determined with very different degrees of pre- 
cision, Mr, Pickering has given them all equal weight, and this 
1s the fatal defect which, it seems to me, entirely vitiates the con- 
clusion reachéd by him. A reference to fig. 1 of my paper in the 
uly number, will show that for the earlier dates, from 1854-0 
down to 1872°6 the graphically determined dates differ from 
the real dates when the record was actually lowered, by from 
One to two years. It will also be seen that the dates 1878°3 
1881-0 are subject to errors which may be as great as two 
€ars, 
After having made a preliminary examination, these dates 
might indeed have been “ adjusted,” so as to make them agree 
tter with the others, and this without giving a “cooked” 
appearance to the reduction; but they now stand exactly as 
they did when first determined and before any. other work ha 
been done. I have plotted the new values of aT with the 
values of s, and the line representing the values so as to give 
most weight to the best determined values, I find to be repre- 
Sented by the equation 
ds 
—=—1° 0°0127 8. 
T 1°24+ 
