STATIC AND GROWTH ALLOMETRY OF MORPHOMETRIC TRAITS IN RANDOMBRED HOUSE MICE.

Abstract

It is a common observation that individuals within a given species tend to develop differences in form or shape during ontogeny, these resulting chiefly from varied rates of growth in a host of morphometric characters (Katz, 1980). Documentation of these differences is important for some aspects of animal and plant breeding as well as for our understanding of evolutionary processes such as selection and adaptation. Quantification of relative growth generally has been accomplished through the use of the simple allometric equation, Y = aXk, or more commonly, its logarithmic equivalent: log Y = k log X + log a. Here Y is the value for a character, X is the value for another character, log a is the Y-intercept, and k is the allometric coefficient. K is known as a static allometric coefficient when it is estimated from the variation in size and shape among individuals all of the same age, as distinguished from a growth k estimated from variation among separate age groups. Whether applied to static or to growth data, the two parameters of the allometric equation have proven to be quite effective empirical descriptors of the variation in growth and form in a wide variety of organisms (Cock, 1966; Gould, 1966; 1971). The very volume of the allometric literature itself, however, has resulted in some confusion both among the rather numerous methods of estimating allometric coefficients, as well as the various kinds or levels of allometry which can be assessed. Although various studies have tended to show that no single allometric estimator is generally the best (Kidwell and Chase, 1967), certain estimators may be preferable in specific applications (Manaster and Manaster, 1975; Smith, 1980). General congruence is probably to be expected among allometric coefficients calculated by various methods if all are derived from the same suite of characters in the same population, but with such an approach, differences might be apparent as well. Studies comparing the congruence of different types of allometry, especially static versus growth, perhaps are even more critical, for our knowledge of the relationship among such types is still rather superficial. Static and growth allometry, for example, have long been assumed to yield parallel results (Simpson et al., 1960), but in spite of the general positive association often found between these two types of allometry (Cochard, 1981; Cheverud, 1982), there seems no reason to expect them to be perfectly associated (Cock, 1966). The principal purpose of this paper is to present and compare various estimates of both static and growth allometric coefficients for a battery of morphometric characters in house mice of three separate ages. Comparisons are made among the different allometric estimators, as well as between the actual patterns of static and growth allometry for these characters. Statistically significant congruence is found for both sets of comparisons, but the extent and reasons for the differences are particularly stressed.