program was condition score (P < .05). Average score for Pro-
gram 1 was 10.9, compared with 11.1 for each of the other two
programs. This difference was too small to be of any economic
significance. For production per cow (a composite of weaning
rate and weaning weight) Program 3 ranked highest with 463
pounds, followed by Program 1 with 445 pounds and Program
2 with 437 pounds. These differences reflect mainly small differ-
ences in weaning rate. The programs ranked in the same order,
but with larger differences, for calf production per acre. There
is no satisfactory test of significance for these differences, al-
though the low value for Program 2 would appear to be real
because of the reduced carrying capacity in the program due to
renovation during the first four years of Phase 3.
 These results are consistent with those from the previous
phase in indicating that Program 1, fertilized at the rate of 300
pounds of 0-10-20 per acre, continued to produce at a high, and
easily the most economical, rate of any program tested. The
clipping plot yields reported in the Agronomy section for Phase
3 are in agreement with the animal response data. In retrospect
it was an error in judgement not to have included also a lower
level of fertilization on clover pastures in an attempt to establish
the point at which further reduction lowered production under
continuous grazing where most of the fertilizer nutrients are re-
turned to the soil.
 The effects of mating systems.-The major items of interest
with respect to the mating systems investigated include: 1. the
comparative performance of foundation animals and their pres-
ent-day descendents, 2. the effectiveness of different mating sys-
tems, and 3. the differential response of different breed groups
on the different pasture programs.
 A progressive improvement in production performance of the
cattle over time is shown in Table 13. These data show that im-
provement occurred for all production traits, with annual pro-
duction per cow increasing from 225 to 448 pounds. A portion of
this increase can be explained by the use of silage for winter feed
and inclusion of clover in more of the pastures. There was
enough overlap of successive generations, however, that the evi-
dence is clear that most of the improvement observed was ge-
netic. The partitioning of effects into genetic and environmental
components will be the subject of .a more detailed publication
elsewhere.
 Differences in the production characteristics of the cattle re-
sulting from the five mating systems are shown in Table 14.