500 



SCIENCE. 



[N. S. Vol. XXII. No. 564. 



been greatly strengthened by the closely op- 

 posed feathers. For several minntes the hawk 

 thus flew alongside of the boat, with quite 

 regular periods of flapping and soaring; then, 

 suddenly shifting its course, it circled out, 

 soaring, passing over my head at a distance 

 of about twenty feet. I could then see plainly 

 that the primaries of one wing (right) were 

 interlocked — the condition of the other wing 

 I had not time to observe. 



My conclusion, therefore, is that the inter- 

 locking of the primaries of hawks takes place, 

 as Mr. Trowbridge has shown, under the con- 

 ditions of soaring in the face of a strong wind. 



Bashford Dean. 



RiNKAT JlKENJO, MlSAKI-MlURA, JAPAN, 



September 3, 1903. 



SPECIAL ARTICLED. 

 THE CHROMOSOMES IN RELATION TO THE DETER- 

 MINATION OF SEX IN INSECTS. 



Material procured during the past summer 

 demonstrates with great clearness that the 

 sexes of Hemiptera show constant and char- 

 acteristic differences in the chromosome 

 groups, which are of such a nature as to leave 

 no doubt that a definite connection of some 

 kind between the chromosomes and the de- 

 termination of sex exists in these animals. 

 These differences are of two types. In one of 

 these, the cells of the female possess one more 

 chromosome than those of the male; in the 

 other, both sexes possess the same number of 

 chromosomes, but one of the chromosomes in 

 the male is much smaller than the correspond- 

 ing one in the female (which is in agreement 

 with the observations of Stevens on the beetle 

 Tenehrio). These types may conveniently be 

 designated as A and B, respectively. The 

 essential facts have been determined in three 

 genera of each type, namely, (type A) Pro- 

 tenor helfragei, Anasa tristis and Alydus 

 pilosulus, and (type B) Lygceus turcicus, 

 Euschistus fissilis and Ccenus delius. The 

 chromosome groups have been examined in 

 the dividing oogonia and ovarian follicle cells 

 of the female and in the dividing spermato- 

 gonia and investing cells of the testis in case 

 of the male. 



Type A includes those forms in which (as 



has been known since Henking's paper of 

 1890 on Pyrrochoris) the spermatozoa are of 

 two classes, one of which contains one more 

 chromosome (the so-called ' accessory ' or 

 heterotropic chromosome) than the other. In 

 this type the somatic number of chromosomes 

 in the female is an even one, while the somatic 

 number in the male is one less (hence an odd 

 number) the actual numbers being in Protenor 

 and Alydus $ 14, 3 13, and in Anasa 5 22, 

 c? 21. A study of the chromosome groups in 

 the two sexes brings out the following addi- 

 tional facts. In the cells of the female all 

 the chromosomes may be arranged two by two 

 to form pairs, each consisting of two chromo- 

 somes of equal size, as is most obvious in the 

 beautiful chromosome groups of Protenor, 

 where the size differences of the chromosomes 

 are very marked. In the male all the chro- 

 mosomes may be thus symmetrically paired 

 with the exception of one which is without a 

 mate. This chromosome is the ' accessory ' 

 or heterotropic one; and it is a consequence 

 of its unpaired character that it passes into 

 only one half of the spermatozoa. 



In type B all of the spermatozoa contain 

 the same number of chromosomes (half the 

 somatic number in both sexes), but they are, 

 nevertheless, of two classes, one of which con- 

 tains a large and one a small ' idiochromo- 

 some.' Both sexes have the same somatic 

 number of chromosomes (fourteen in the three 

 examples mentioned above), but differ as fol- 

 lows : In the cells of the female (oogonia 

 and follicle-cells) all of the chromosomes may, 

 as in type A, be arranged two by two in equal 

 pairs, and a small idiochromosome is not pres- 

 ent. In the cells of the male all but two may 

 be thus equally paired. These two are the 

 unequal idiochromosomes, and during the 

 maturation process they are so distributed 

 that the small one passes into one half of the 

 spermatozoa, the large one into the other half. 



These facts admit, I believe, of but one 

 interpretation. Since all of the chromosomes 

 in the female (oogonia) may be symmetrically 

 paired, there can be no doubt that synapsis in 

 this sex gives rise to the reduced number of 

 symmetrical bivalents, and that consequently 



