An analysis of muscle weight variations in the wing and leg of Sturnus vulgaris.

Dry weights of 18 wing and 23 leg muscles from both sides of 15 adult male starlings (Sturnus vulgaris) were determined. By means of factor analysis it is calculated which part of the variance in a muscle's weight can be accounted for by common factors and which part by the characteristics of the muscle itself (residual variance). Wing and leg muscles are analysed separately. From the wing muscle analysis one can draw the following inferences: 1. Three hierarchically ordered factors represent 66%, 8% and 6% of the total variance in the wing muscles. The first common factor will most likely reflect the size variation of the starlings. 2. It appeared advisable to perform rotation of the original factor solution because the second and third factor are bipolar. 3. The three rotated factors can not be interpreted as embryological sources of weight variations and are not related to the segmental or peripheral innervation of the muscles, but generally correspond with three groups of actions that the wing muscles perform. 4. Muscles correlated with the first rotated factor generally move the humerus, while muscles that move the forearm and (parts of) the hand are correlated with the third and second rotated factor, respectively. From the leg muscle analysis it follows that: 1. Five hierarchically ordered factors represent 57%, 13%, 7%, 6% and 4% of the total variance in the leg muscles. Body size (factor 1) plays a more important role in the wing than it does in the leg. 2. It was advisable to rotate the original factor solution since four bipolar factors are present. 3. As in the wing, residual variance shows no particular relation with the mean weight of a muscle. 4. As in the wing, the five rotated factors can only be related to five groups of muscle actions. 5. Muscles correlated with the five rotated factor cause protraction, adduction or inward rotation of the leg or combinations of these movements, whereas muscles correlated with the fourth rotated factor can perform the antagonistic actions. Muscles related to the third rotated factor flex the tarsometatarsus of flex more digits simultaneously, while muscles related to the second rotated factor flex a single digit or extend the tarsometatarus. The fifth rotated factor is a peculiar one since only the mm. peronei are correlated with it. The possibility to apply factor analysis also to experimental stimulated hypertrophy studies and to interspecific muscle weight analyses is discussed.