Mean annual egg production It is comparatively easy to arrange the annual averages and proportions describing these aspects of egg production in a sequence and to determine whether the direction of the sequence is upward, downward or stationary. It is less easy to interpret and estimate the significance of results so obtained for any single breed. Caution is necessary because of the small groups of fowls on which our averages are founded in some of the years and because the conditions in all of the years are not strictly comparable. After making corrections for some of these factors, however, a good description of the changes may be obtained, while a consideration of the causes involved may be deferred until data from all of the breeds are considered together. Fig. 2. Changes in mean annual egg production 1911-1919. The solid line AB shows the trend of changes in the mean through the nine year period. The broken line CD shows the trend of changes through the seven normal years designated in italics. In order to determine roughly the direction of changes which have taken place in average annual egg production, we have plotted the annual means from Table 2 in Fig. 2. The course of the changes in mean production is shown by the zigzag solid line in this graph. The solid straight line (AB) which passes through this irregular series of points is the straight line which most nearly fits the actual means, as found by the method of least squares. The equation to this line is y=143.130+3.301x. Its slope is quite definitely upward indi OF THE DOMESTIC FOWL cating a decided tendency for egg production to increase through the nine years. The indicated increase if it were distributed equally over the whole nine years would be about 3.3 eggs per year, and the total increase in nine years about thirty eggs. The actual means of the nine years, however, show that the increase was not as gradual as the fitted line might indicate, but actually rather abrupt, the really significant increase taking place after 1915. Such a sudden change in egg production has not been noted in the other breeds and its causes are probably to be sought in some rather abrupt change Fig. 3. Annual change in the percentage of pullets laying 210 eggs per year and more. The solid line AB shows the trend of changes in the proportion of low producers through the nine years. The broken line CD shows the trend of changes in the seven normal years designated in italics (1913 and 1918 omitted). in the kinds of Barred Rocks submitted rather than in changes in the environment under which they made their records. Omitting those years in which the conditions are known to have been poor (1913 and 1918) and fitting a straight line (CD in Fig. 2) to the means for the seven normal years as a sequence increases somewhat the upward slope of the line. The equation to this seven-year line is y=141.567+4.983x and the total theoretical increase in mean egg production is about forty-five eggs for the seven years. If a line is fitted to the seven normal years retained in their original positions (1, 2, 4, 5, 6, 7, 9) the tendency toward increase is still evident, since the equation to such a line is y=143.399+3.727x. The results from all three methods of calculation agree in indicating that the average egg production of the Barred Rocks in these contests has increased considerably. In spite of the increase in mean production, the fowls entered from year to year have not become more uniform in egg production. The equation to the line describing changes in the coefficient of variation (y=26.065+.170x) indicates that the variability of the fowls in the nine years has tended to be about the same. B 1911 1911 1912 1913 1914 1915 1916 1917 1918 1919 Fig. 4. Annual changes in the percentage of pullets laying 104 eggs per year and less. The solid line AB shows the trend of changes in the proportion of high producers through the nine year period. The broken line CD shows the trend of changes through the seven normal years designated in italics, (1913 and 1918 omitted). Th distribution of the fowls in regard to egg production have, however, shown some changes in a fairly constant direction. This is most easily demonstrated by calculating the proportions of fowls which in each year lay many and few eggs, as measured by arbitrary standards of production. The proportion of each annual population which lays 210 eggs and over, is compared in Table 4 with the proportion which lays 104 eggs and less. The direction of changes in these propor OF THE DOMESTIC FOWL tions is shown in the straight lines in Figs. 3 and 4. The slope of the nine year line (AB Fig. 3) describing changes in the proportion of birds laying 210 eggs and over is given by the equation y=.422+2.240x; for the line CD which is founded on the proportions for the seven normal years only (as a sequence) the equation is y=.014+3.279x. In either case a distinct tendency is in evidence toward increase in the higher producing classes of fowls. With regard to the proportion of low producers (Fig. 4) the tendency while apparently opposite in direction is much less marked. The fitted line AB slopes downward, but its equation (y-12.781-585x) indicates that there has been only a slight tendency for the proportion of poor producers to decrease. Omission of the low years (1913 and 1918) each of which is characterized by a large proportion of low producing fowls,, accentuates the downward slope of the line (shown by CD, Fig. 4, equation y=11.971-825x) but leaves it less steep than the line showing the increase in the proportion of high producers. Since the relative numbers of high producing fowls has tended to increase markedly, while the low producers have declined but little, the variability of the annual groups has tended to remain rather high. These changes as we have seen are dependent not only on the kinds of fowls submitted, but also on the environmental conditions encountered in different years, since in years of poor conditions the proportion of high producing fowls is low and the numbers of low producing fowls are large. We have partially corrected for those changes in conditions by omitting the years of poor conditions. But if we wish to detect changes in the relative frequency of innately good and poor producers, we must define high and low egg production as relative to the conditions under which it occurs. This may be done by defining high producers as those which exceed the average production of the group by a given percentage, and poor producers as those which are below the mean by a like percentage. In this case, we have determined the annual proportions of fowls which lay 33 1/3 percent more and 33 1/3 percent fewer than the average number of eggs. These proprotions are, there fore, dependant on the average for the year in which they occur and provide a truer index of the numbers of relatively superior and inferior individuals in each population. The proportions of such fowls for each year is given in Table 4. It is obvious that high and low production so defined is more or less independent of environmental conditions, since the low years 1913 and 1918 are not marked by the highest proportions of low producers or the lowest proportions of high producers. When straight lines are fitted to these proportions it is found that neither has shown any significant tendency to change during the nine years. Thus the line for the low producing fowls has the equation y-10.866-.093x, while for the high producers it is y=10.818-.168x. The proportions of fowls in the upper and lower ends of the annual distributions have tended to remain the same through the nine years; and the improvement in mean production and in the numbers of fowls which surpass an arbitrary standard has not been effected through a change in the essential form of the distribution of egg production. We have been unable to find other contest data on Barred Rocks for a similar succession of years so that we cannot check and compare our conclusions on the tendency toward changes in egg production in this breed with changes in other places. THE SEASONAL DISTRIBUTION OF EGG PRODUCTION. The second part of the statistical study of egg production in each breed for which there is sufficient material in the contest records is devoted to an analysis of individual variation in egg production in the months and seasons of the year. In the present case this will consist of a discussion of the average number and proportion of eggs laid by Barred Rocks in each month and season of the year: followed by a description of the seasonal cycles of egg production found in this breed, and finally will attempt to discover whether the seasonal distribution of egg production has shown any significant tendency to change during the nine years which the records cover. |