Three methods of estimating the mean shell thickness of a single egg are considered in detail: by micrometer, by shell weight per unit area and by shell deformation. Descriptions are given of new variants of these methods. An experiment is described in which the error of estimation of twelve methods was measured. The precision of the deformation method can be improved by measuring the deformation across the poles as well as the equator and using a weighted mean. It can be improved further by supplementing the deformation measurements with measurements of shell curvature.
The problem of estimating flock mean shell thickness differs fundamentally from that of estimating the mean for a single egg. An experiment is described in which the between‐eggs‐within‐hens and between‐hens‐within‐flocks components of variance of shell thickness were estimated, and estimates are given of the standard error of the flock mean obtained by the twelve methods. Four simplified, approximate methods are described.
There are two main purposes for which it is necessary to estimate the mean shell thickness of a single egg: for scientific research and to identify individual thin‐shelled eggs in packing stations. Flock mean shell thickness requires estimating for three main purposes: selection in breeding programmes, diagnosis in commercial flocks and identifying thin‐shelled consignments in packing stations. The purpose for which the estimate is required determines the choice of method.