The paper explores the possibility of implementing the methods of multivariate statistics into studying the growth processes on the example of song thrushes, Turdus philomelos, Brehm, 1831, during their postnatal development as nestlings. The developmental trends in 12 morphometric traits in T. philomelos in the course of postembryogenesis is shown to be explained for 99.3 % by the first two principal components (PC). The major developmental trend (PC1 - 95.1 %) is defined by a highly correlative though irregular growth of linear forms of nestlings’ body parts, the two other trends relate to the body proportion formation (PC2 - 4.2 %). Th ere have been discovered the two growth stages: (1) of fast growth: from birth up to the 8th day with relative increment in growth of traits equal in average to 91.9 %, and (2) of slow growth: from the 8th to the 14th day, characterized by a reduction of an average growth increment being five times lower, and by intense feather cover development. There have been demonstrated that all the variables can be structured into the four groups or growth correlation pleiads (groups comprising similarly growing traits). The growth is shown to be most specific for the song thrush’s body, head and bill, being a part of a single pleiad. While still forming the three different growth pleiads the properties of the bird’s wing, leg, the 3rd and the 4th toes differ significantly less in respect to their growth characteristics.