The proportion between the mean number of eggs actually laid and the total possible number is given in Table 8. The significant column for our purpose is that giving the percentage increase of production during the winter months. This gives a truer picture of the actual circumstances than the means for the various months, because the variation in the number of days in the several months has been eliminated by reducing all to a percentage basis. This is especially noticeable in the case of February where the mean production is based on only 28 laying days in 6 years and 29 days in three years (average 28.33 days). The increase in rate of November production is probably not as high as given, since some eggs have doubtless been laid before the opening of the contest. The increase of December over November may likewise be too high, because of the delay in beginning egg production as an after effect of shipment of the birds. By January this effect has probably ceased to operate and the rate of interest in January has slackened. This represents only a decline in the rate of increase, not a decline in mean production. It may be an indication of a winter cycle which would then end a full month before the natural spring breeding season begins. I am inclined to think, however, that the lessened rate in January is due rather to an exaggeration of the delayed December increase. Certainly the increase of March over February is not significantly different from the increase of February over January; and if these Wyandottes resembled the Barred Rocks in this respect, it is between these two months, February and March, that we should find our clearest evidence of the existence of a winter cycle. The absence of such evidence indicates that in these contest Wyandottes the existence of a winter cycle cannot be affirmed. The above data are treated en masse, and the conclusions are limited to the whole number of birds considered as a group. The dissection of this groap into its component individuals might throw some light on the existence of a winter cycle in certain individuals. Such a minute analysis will not be attemped here. We have instead separated the birds into the 21 fecundity groups used in the seriation of the annual distributions. These differ by 15 eggs each, and the distribution of egg laying in the various months is shown for each group in Table 9. This table is discussed in greater detail elsewhere. At present, we wish to know whether these groups differ one from another in respect to a winter cycle. Such a difference should probably be most apparent between high producers (210 eggs and over) and low producers (104 eggs and under). This is probable, because we know that winter production is an anomaly in wild fowls and has been developed arti icially by selection of extreme variates (or of mutants) in this direction, and because the chief increases in fecundity have been additions to seasons other than the normal season of egg laying in the spring. The seasonal distribution of egg production in low and high producers is shown in Table 10 and Fig. 7. Noting at present only the winter months, we cannot fail to remark the very great differences between the classes in the percentage of the total yearly production which occurs at this period. The rate of production is likewise different, being steadily upward in the low producers, while in the high producers the large increase in December over November production is followed by a very slow increase in January and February. The increase of February over January is in fact probably not significant. March shows a sudden increase in both classes marking the opening of the spring period. The uniform rate of production of the high producers is in marked contrast with the rising rate of the low producers. No let-up in this rate occurs to indicate the looked for pause between the winter and spring production. If we follow the percentage production in the winter months down through the various fecundity classes in Table 9, we find that the changes from the "low" type of winter production to the "high" type are continuous and gradual. High winter production is not attained by a discontinuous change such as we might expect if it were due to the addition intact of a separate and distinct cycle to the laying year. As far as the evidence from these Wyandottes is concerned, therefore, we must conclude that evidence of a distinct cycle is lacking, when all the birds are considered together or when the group is broken up into smaller classes differing in annual fecundity. 2. Spring Period. The usual spring cycle of production is undoubtedly exhibited by these Wyandottes. This cycle in fact is probably present in all breeds of poultry and species of wild birds, since upon fecundity at this, the mating time, depends the reproduction and hence the survival of the species. Its inception is attended by an increase in mean production of March over February; a decrease in variability and the fact that practically every bird in the flock is laying in March. It is, however, defined better by the general character of egg production during the whole period than by conditions surrounding its beginning or its ending, since we have seen that by all the criteria applied, the change from winter production into spring production is a gradual one. The winter period of fecundity was a time of increasing egg production; the spring is a period of stationary production. It is a climax of which the winter production was a foreshadowing, and the subsequent production an aftermath. In the case of these Wyandottes, this reproductive climax is extraordinarily sustained, since production is maintained at the same level through three months, March, April, and May. 3. Summer Period. The ending of the spring or normal reproductive cycle is as poorly defined as its beginning. Ordinarily, broodiness ensues after the heavy spring production and a marked reduction in laying activities is noticeable. Conventionally the last of May is taken to mark the close of the reproductive cycle and the opening of the summer period. In the Wyandottes, however, this cessation is delayed and egg production in June is very similar in mean and variability, to egg production in the spring months. After June, egg production reaches a new stationary level lasting through July, August, and September. June may then be considered, on the basis of fecundity either as a part of the spring or summer cycle. The truth is undoubtedly that egg laying in this and in the summer months generally is not a result of a separate or induced physiological cycle, but merely a continuation of the one chief spring production period. The mechanism of reproduction, having been wound up, is only slowly running down. One factor in the running down in mean production after the spring cycle is the onset of broodiness and the cessation of laying on the part of broody birds. This has not had a serious effect on summer production in Wyandottes for a high mean production is maintained and the number of zero producers (in this case usually birds which have stopped laying) is relatively small-2.7 per cent of the flock in June, 4.1 per cent in July, and 4.5 per cent in August (see Table 11). The slight increase in mean production in August, which has been already noted, probably indicates not only the maintenance of high egg production which is the striking feature of summer production, but also the recovery from broodiness on the part of a few affected birds. If the characteristic of autumn production be the entrance of many of the birds into a period of reproductive quiescence accompanied by a molt, then in the Wyandottes this period is represented by but one month, October. September production differs only slightly from August in the mean, in variability and in numbers of birds which have ceased laying. September belongs more properly, then, with the summer months in that egg production is maintained and nearly all the birds are laying. The change from September to October is, however, sharp. Mean egg production falls from 14.06 to 10.07; the coefficient of variability rises sharply from 50.10 to 75.09 and the percentage of the flock not laying rises from 6.7 per cent to 20.7 per cent (see Table 7). The onset of moulting in so far as it can be gauged by egg production is, therefore, relatively late in Wyandottes. They are good summer and fall producers and their leading position among the breeds submitted to the contest probably owes much to their superiority in these periods. ANALYSIS OF EGG PRODUCTION IN THE SEPARATE MONTHS No attempt has been made to analyse exhaustively the frequency distributions of the separate months, since from the tabled frequencies (Appendix Table II) a close resemblance can be traced to the distributions for Barred Rocks which have been carefully analysed and illustrated by Pearl and Surface. In general the frequency distribution for each month is bimodal, indicating a division of all the birds in any one month into two groups. One group, with a modal production of no eggs (or very few eggs) may be called the zero (or low) producers. In the months before March these are chiefly birds |