66 THE INDIAN JOURNAL OF VETERINARY SCIENCE AND ANIMAL HUSBANDRY [ I, II

                           MATERIAL AND METHODS.

         On all the Government dairy farms in India, two classes of animals have been
maintained, pure Indian and cross-breds. Of the pure Indian stock, the best and
most well managed herds and which consequently give the most reliable data for
such a study as this are the pure-bred Sahiwal herds at the Military Dairy Farms,
Lahore and Ferozepur, and at the Imperial Institute of Agricultural Research, Pusa.
Hence for the study of Indian cattle the milk records of the animals from these
herds only have been utilized.

         But by far the greatest number of animals bred and kept on these farms is of
cross-breds. The milk records of the Military Dairy Farms' cross-breds, however,
suffer from a very serious defect, for most of these animals go on moving from place
to place along with troops, and consequently the milk-records produced by them
are affected by another very potent factor hard to allow for, i.e., change of locality
and therefore climate, etc. None of these is therefore used in studying the effect of
variation of these factors. The records of the Pusa cross-bred animals (1/2
Ayrshire-Sahiwal) which are free from this defect have been separately used in
studying the effect of age on lactation milk yield in that breed.

         Only normal lactation milk records were utilized, all lactations being considered
abnormal and therefore excluded in which :—

         (1)  the cow failed to conceive again, and therefore in which S. P. could not
                  be found ;

         (2)  the calf was allowed to suck the cow ;

         (3)  there was any serious illness ;
                  and

         (4)  the cow aborted.

         1496 complete lactation milk records from pure bred Sahiwals and 348 from
cross-breds were obtained and have been utilized for this study. Since Sanders'
work has shown that the effect of the season of the year varies from district to
district along with variations in herd management, only Lahore and Ferozepore
records—1274 in number—were used for this factor. It is much regretted that due
to the paucity and incompleteness of data in several respects the study of this
factor could not be very exhaustive. Though the results obtained may not be quite
conclusive, yet it may be said that a presumption has been laid that no such effect
of the season of the year on milk yield seems to exist in our data as reported by
Sanders. For the study of other factors all the 1496 records have been used. The
terms season of the year, service period, age, and dry period have the same
meaning as implied in Sanders' work.