138 Dr. Ct. C. Crampton on the Primitive 



connection, tliat I have dissected specimens of Termopsis 

 ampisticollis Hagen, which are placed in the family Proter- 

 mitidae (to which Mastotermes also belongs) by Holmgren, 

 and an examination of these forms has served to substantiate 

 the conclusions here drawn concerning the sexes of the 

 alate caste of Mastotermes. 



In both types of winged forms of Mastotennes (PI. IV, figs. 

 1 and 2), there are ten visible tergal plates in the abdomen. 

 The sternal plates, however, are not situated immediately 

 below their corresponding tergal plates in the posterior 

 region of the abdomen, and the number of sternal plates is 

 not the same in the two sexes, there being but six apparent, 

 well-developed, pigmented, sternal plates in the winged 

 forms which I have interpreted as the females, while there 

 are eight of these sternal plates in the winged " males." 

 In both forms the sternal plate of the real first abdominal 

 segment has become atrophied (or at any rate, it cannot 

 be readily detected), so that what appears to be the first 

 abdominal sternite, is in reality the sternite of the actual 

 secoiid abdominal segment, while what appears to be the 

 second sternite, is in reality the sternite of the actual third 

 abdomijial segment, and so on. 



As was mentioned above, there are apparently but six 

 distinct ventral plates in the abdomen of the winged 

 " female " of Mastotermes, and since what appears to be 

 the first sternite is in reality the sternite of the actual second 

 abdominal segment, etc., the apparent sixth ventral plate 

 (" hg " of PI. IV, figs. 1 and 3), which is unusually large, repre- 

 sents the sternite of the actual seventh abdominal segment. 

 This is in agreement with the statement made by Holmgren, 

 1909, on page 150 of his " Terraitenstudien," that the 

 seventh sternite is much larger than the others in the 

 abdomen of female termites in general. 



I at first thought that the stip])led terminal area of the 

 seventh abdominal sternite shown in PI. IV, figs. 1 and 3, 

 might represent the remains of another abdominal sternite 

 entering into the composition of the unusually large sternite 

 " hg," since in the alate females of our Californian Proto- 

 termitid Termo])sis (which are more primitive than those of 

 Mastotermes in having retained a distinct sternite behind 

 the seventh abdominal sternite) a distinct, though small, 

 eighth sternite occurs in approximately the same position 

 as that occupied by the terminal stippled area of the sternite 

 labelled " hg " in PL IV, figs. 1 and 3. The condition 



