HAROLD HEATH AND BLAKE C. WILBUR. 



dominal wall due to food in the digestive tract and the develop- 

 ment of the fat body, and in the later stages of this first instar, 

 when the insect is about to molt, the body occasionally may become 

 even larger than the smallest individuals of the succeeding instar. 

 Body length, therefore, is not an absolutely dependable criterion 

 for the determination of a given instar. Considerably better is the 

 customary method of determining instars on the basis of antennal 

 segments. This likewise is not always a trustworthy guide, espe- 

 cially in the later stages ; but it has been found that where these 

 antennal joint counts are combined with the measurement of the 

 transverse diameter of the head the difficulties of distinguishing 

 the various stages practically disappear. 



In recently hatched young of T. nevadensis the head width 

 averages 0.648 mm. on the basis of 182 individuals taken from 

 25 colonies the minimum being 0.58 mm. and the maximum 0.67 

 mm. The number of antennal joints is usually thirteen though 

 twelve may occur. In some specimens this variation may appear 

 on opposite sides of the same individual. The long third joint 

 appears to be unsegmented, but in specimens treated with caustic 

 potash and stained with magenta this segment is found to be sub- 

 divided into two incipient segments of which the distal one is the 

 smaller. The first instar in the case of T. angusticollis is prac- 

 tically a duplicate of the same stage in T. nevadensis. The trans- 

 verse head diameter, however, is somewhat greater. In 116 in- 

 dividuals from twenty-four separate colonies the maximum width 

 was o.Si mm., the minimum 0.74 mm. with an average of 0.7 mm. 

 The number of antennal segments is fourteen. 



The distribution of the spines over the body of recently hatched 

 individuals in young colonies of T. nevadensis is shown in PI. I, 

 Fig. i ; of T. angusticollis in PI. II, Fig. 7. Even in the oldest 

 communities, headed by numerous substitute royal insects, there 

 are no appreciable differences where individuals of the first instar 

 are concerned. In other words the newly hatched young of each 

 species are alike externally whether they belong to a recently es- 

 tablished community or one that has existed for several years. 

 An exception to this rule may be noted in the case of young 

 hatched in colonies headed by old primary royal pairs or by sub- 

 stitute royal forms. In such cases the spines show the same gen- 



