ON THE RELATION BETWEEN LITTER SIZE,BIRTH WEIGHT,AND RATE OF GROWTH,IN MICE

We have been concerned with the connection between size of litter and weight of litter at birth, especially in mice. The weight at birth represents, it is to be presumed (at least in mice, and for certain other cases), the weight at a particular developmental stage. The connection between number in litter (N) and weight of litter (W) has been interpreted as due to the partition of nourishment between mother and young, and on an equal basis among the several embryos of a litter. The "heterogonic" relationship which the data exhibit between N and W shows that the constant K, defined by log W = K log N + const., is independent of the species, and has an essentially constant value (0.85±) in all multiparous mammals; it is therefore regarded as a partition coefficient. In the case of power function relationships between masses of components of a single individual, the respective "drawing powers" of the several organs are diverse, and diverse magnitudes of K are encountered. With developing embryos, the intrinsic drawing powers of the tissues concerned in embryos and mothers are in each case of the same general character, at least among mammals; the constancy of K reflects this. A parallel for the case as it appears in the consideration of relative growth rates of organs in a single individual, and in which the varying magnitudes of the heterogonic growth constant K are presumed to reflect diverse drawing powers of the respective tissues, would be given by intrauterine growth of a litter containing individuals with diverse capacities for growth, —that is, individuals differing genetically with respect to the factors determining the magnitudes of w ₁. We have been dealing with the growth of litters in inbred strains. It is to be presumed that in the case of the growth of a litter containing two categories of individuals so far as concerns intrinsic drawing powers with respect to the nourishment provided by the mother, it would be possible to investigate the way in which K is open to modification. Although difficult, from the standpoint of classifying the individual young, it would appear to be distinctly worth while to make such an experiment, and we have planned it for the future. It is pointed out that for genetic purposes the ideal weight of a litter of 1 is obtainable from a series of measurements of N and W, free from disturbances affecting the apparent value of this quantity as observed in single births. This weight of an ideal litter of 1 should be employed to disentangle the effects of heterosis and fertility factors from those having to do with individual weight at birth. During the suckling period the relation ΔW/W = K (ΔN/N) is maintained for young mice, but with modifications in the case of small and large suckling litters due to (1) the time course of milk yield, and (2) the effect of litter size upon this. It is shown that a growth curve can be obtained for an ideal litter of I, under the condition of milk supply that on each day the mother is able to provide a constant fractional increase of milk for each additional young mouse in the litter. The rate of growth then adheres to the time curve of capacity for production of milk.