PERFORMANCE OF MORGAN HORSES UNDER SADDLE 33 
It should be borne in mind that the primary purpose of the breeding 
work at the United States Morgan Horse Farm has been to produce 
animals for use both under saddle and in the carriage. To qualify as 
good saddle horses, they should have clean gaits that are easy on the 
rider, be easy to handle, possess considerable endurance, and have a 
reasonable speed. Excessive speed, as that of race horses, is not 
desired or attempted. 
The speed that a horse attains in a given test is due to the influence 
of many factors. From a breeding standpoint, the factors that ap- 
peared to be most important in this study were length of stride, 
temperament, action, length of leg, slope of shoulder, and slope of 
croup. Shape and size of feet, which were found to be associated 
with speed of walk, were probably coincidental combinations. The 
five horses with the smallest feet were by two sires, Canfield and Gold- 
field, which also had the fastest walking progeny. Sex undoubtedly 
had an influence on speed both directly and through its association 
with length of stride at the trot and general conformation. Unfor- 
tunately, there was not a normal distribution of sexes. The castra- 
tion of most of the males and disposal of others before testing were 
undoubtedly factors that resulted in the elimination of certain types. 
The fact that the stallions, on the average, had a shorter stride at the 
trot than the geldings might indicate a selection for a type with a 
shorter stride. On the other hand, it might be due to a difference 
in temperament directly connected with the: presence or absence of 
the sex glands. 
At the walk, a long stride was not necessarily associated with speed 
since a considerable number of horses with long strides did not 
naturally walk fast. Speed, however, was dependent on the horses’ 
having a reasonably long stride since none with short strides walked 
fast. Since straight, snappy action was associated with speed at the 
walk, it is evident that ability and willingness to move the feet fast in 
combination with ability and willingness to take a long stride are 
essential for a horse to make good time at the walk. Although 
slope of shoulder, which was found to be associated with time required 
to walk a mile, did not seem to be associated with length of stride, it 
was associated with action or willingness and ability to move the 
feet fast. 
At the trot, none of the horses with a long stride were very slow 
and none with a short stride were fast. Slope of croup was also 
associated with speed at the trot but was not significantly associated 
with length of stride. The apparent lack of association where it 
might be expected may be partly due to the fact that medium slope 
of croup was found to be the most favorable. The same statement 
may apply also to slope of shoulder in relation to speed of walk, and 
to top line of back and withers where a medium condition was associ- 
ated with a longer stride at the trot but not with the greatest speed 
at the trot. Also the lack of association may be due to the characters’ 
being related to speed in different ways but not necessarily to one 
another. 
On the cross-country ride, speed was an even more complicated 
measure since it involved three gaits and, furthermore, the long 
distance might involve endurance to a much greater extent than the 
1-mile tests. The fact that only seven horses showed marked ‘signs 
