No. 373.] REVIEWS OF RECENT LITERATURE. 57 
trotters and 122 pacers trace their descent in one or more lines from 
this horse. The preponderance of pacers is accounted for by the 
greater swiftness of their gait. Because of the inherently greater 
speed of the pace over the trot, it will be necessary, in order to com- 
pare the speed attained by a pacer with the speed of his trotting 
ancestors or brothers, to establish some ratio by which a trotting 
record may be transmuted to its equivalent pacing record, in the 
same way that Galton has transmuted female stature into its male 
- equivalent in his discussions of the statistics of human measurements. 
This will require the comparison of a large number of individuals. 
In the meantime, wishing to gain some idea of what this ratio may 
be, we have compared the 54 best pacers with the 54 2:10 trotters. 
Comparing each horse of one class with the horse of the corresponding 
grade in the other, there is found to be an average difference of 2% 
seconds, the maximum being 334 seconds and the minimum 134 sec- 
onds. It is interesting to note in this connection that in the case of ° 
one horse in our list that has made fast records in both classes the 
difference is not more than the above maximum, the pacing record 
of Jay-Eye-See being 2:06, and his trotting record 2:10. If this 
difference represents the gain in speed which a horse equally gifted 
in both gaits would make in pacing, then all horses who can trot 
within 2:1214 should be classed with the 2:10 pacers. At any rate, 
it is unfair to compare 2:10 trotters with 2:10 pacers, and for this 
reason the tables on pages 27 and 28 are misleading. 
The author points out another source of error which arises from 
the introduction of the bicycle sulky with pneumatic tires in 1892. 
But, allowing for errors due to bicycle sulkies, improved tracks, and 
more experienced trainers, we can see a gradual increase of trotting 
and pacing speed in successive generations. How much of this 
improvement is due to the inherited effects of training, and how 
much to selection and combination of favorable variations in breed- 
ing? The list shows that a number of stallions and mares, after 
having been trained to fast records, have got foals that have made 
fast records. But there is no evidence that a line of trained ances- 
tors is more successful in producing speed than a line of untrained 
ancestors, or a line of mixed trained and untrained ancestors. For 
example, of the 122 pacers in the list only 8 have a parent or grand- 
parent that has paced in 2:10 or trotted in 2:13. None of the 50 
trotters has a parent with a 2:10 record. In the list of trotters both 
parents are given in 22 cases. Both parents have a record in only 2 
cases; in 13 Cases one parent only has a record; and in 7 cases neither 
