STUDIES IN SPERMATOGENESIS

PART II.

(Pages 33-74. Plates VIII-XV.)

A COMPARATIVE STUDY OF THE HETEROCHROMOSOMES IN CERTAIN SPECIES OF COLEOPTERA, HEMIPTERA AND LEPIDOPTERA, WITH ESPECIAL REFERENCE TO

SEX DETERMINATION.

BY N. M. STEVENS

WASHINGTON, D. C.:
Published by the Carnegie Institution of Washington
October, 1906

CARNEGIE INSTITUTION OF WASHINGTON

Publication No. 36, Part II.

FROM THE PRESS OF
THE WILKENS-SHEIRY PRINTING CO.
WASHINGTON, D. C.

STUDIES IN SPERMATOGENESIS.—II.

A COMPARATIVE STUDY OF THE HETEROCHROMOSOMES IN CERTAIN SPECIES OF COLEOPTERA, HEMIPTERA, AND LEPIDOPTERA, WITH ESPECIAL REFERENCE TO SEX DETERMINATION.

By N. M. Stevens.

INTRODUCTION.

In Part I of this series of papers, the spermatogenesis of five species belonging to four different orders of insects was considered. In two species of Orthoptera an "accessory chromosome" was found; in Tenebrio molitor, one of the Coleoptera, an unequal pair of chromosomes was described; in the other species no heterochromosomes were discovered. The apparent bearing of the chromosome conditions in Tenebrio molitor on the problem of sex determination has led to a further investigation of the germ cells of the Coleoptera. One of the Hemiptera homoptera and two of the Lepidoptera have also been examined for comparison with the Coleoptera and the Hemiptera heteroptera.

METHODS.

As a result of previous experience with similar material, only two general methods of fixing and staining have been employed: (1) Fixation in Flemming's strong solution or Hermann's platino-aceto-osmic, followed by either Heidenhain's iron-hæmatoxylin or Hermann's safranin-gentian staining method (Arch. f. mikr. Anat. 1889). (2) Fixation after Gilson's mercuro-nitric formula, followed by iron-hæmatoxylin, Delafield's hæmatoxylin and orange G, Auerbach's combination of methyl green and acid fuchsin, or thionin.

The iron-hæmatoxylin with either mode of fixation gives by far the most satisfactory preparations for general study. The other stains were used mainly for the purpose of distinguishing between heterochromosomes and plasmosomes in resting stages of the nucleus.

COLEOPTERA.

Trirhabda virgata (Family Chrysomelidæ).

Two species of Trirhabda were found in larval, pupal, and adult stage on Solidago sempervirens, one at Harpswell, Maine, the other at Woods Hole, Massachusetts. The adult insects of the two species differ slightly in size and color, the germ cells mainly in the number of chromosomes, Trirhabda virgata having 28 and Trirhabda canadense 30 in spermatogonia and somatic cells.

In Trirhabda virgata, the metaphase of a spermatogonial mitosis (plate VIII, fig. 3) contains 28 chromosomes, one of which, as in Tenebrio molitor is very much smaller than any of the others. The maternal homologue of the small chromosome is, as later stages show, one of the largest chromosomes. In Tenebrio the unequal pair could not be distinguished in the growth stages of the spermatocytes. In Trirhabda it has not been detected in the synizesis stage (fig. 4), but in the later growth stages (figs. 5-7) this pair is conspicuous in preparations stained by the various methods cited above, while the spireme is pale and inconspicuous. The size of the heterochromosome pair varies considerably at different times in the growth period, and in some nuclei (fig. 7) both chromosomes appear to be attached to a plasmosome. The ordinary chromosomes assume the form of rings and crosses in the prophase of the first maturation mitosis (fig. 8), but usually appear in the spindle as dumb-bells or occasionally as tetrads (fig. 10), or crosses (fig. 11). The unsymmetrical pair is plainly seen in figures 9 and 11, but is not distinguishable in a polar view of the metaphase (fig. 13). In the anaphase (figs. 14-16) the larger and the smaller components of the pair separate as in Tenebrio. This is, therefore, clearly a reducing division as far as this pair is concerned, and probably for all of the other pairs, though neither the synapsis stage nor the prophase forms are so clear on this point as in some of the other species studied. Figures 17 and 18 show metaphases of the two classes of second spermatocytes, the chromosomes varying somewhat in form in different preparations and even in different cysts of the same preparation. An early anaphase of this mitosis is shown in figure 19; here the small chromosome is already divided. It was impossible to find good polar views of the daughter plates in the two classes of second spermatocytes, but it is evident from figure 19 and other similar views of the second spermatocyte spindle that, as in Tenebrio, one-half of the spermatids will contain one of the derivatives of the small chromosome, the other half one of the products of its larger homologue.

Sections of male pupæ were examined for equatorial plates of somatic mitoses. Figure 1 is a specimen of such plates. As might be expected, this figure resembles quite closely the spermatogonial equatorial plate (fig. 3) in number, form, and size of chromosomes, the small one being present in both. Figure 2 is from the follicle of a young egg; here we find 28 chromosomes, but no small one. The chromosome corresponding to the larger member of the unequal pair in the male evidently has a homologue of equal size in the female. The chromosome relations in the male and female somatic cells are therefore the same as in Tenebrio molitor, and must have been brought about by the development of a male from an egg fertilized by a spermatozoön containing the small chromosome, and a female from an egg fertilized by a spermatozoön containing the larger heterochromosome.

Trirhabda canadense.

In Trirhabda canadense the spermatogonial chromosomes are invariably smaller than in T. virgata, but similar size relations prevail. The spermatogonial plate (fig. 21) contains 30 chromosomes, 29 large and 1 extremely small. In the growth stages the association of the two unequally paired chromosomes with a rather large plasmosome is more evident than in T. virgata (figs. 22-23). In this species the unequal pair is more often found at a different level from the other chromosomes in the early metaphase of the first maturation mitosis (fig. 24), but it later comes into the plate with the other chromosomes (figs. 25-27), and divides earlier than most of the other bivalents (fig. 27). In a polar view of this metaphase the largest chromosome often appears double (fig. 28); in a front view it is a tetrad as in T. virgata, figure 10. Figure 29 is the equatorial plate of a metaphase in which the larger component of the unequal pair has been removed in sectioning. The daughter plates of a first spermatocyte in anaphase (fig. 30) show the separation of the components of the heterochromosome pair; and equatorial plates of the resulting two classes of second spermatocytes (fig. 31) show the same conditions. Figures 32 and 33 are prophases of the second division, figure 33 showing the small chromosome ready for metakinesis. It was impossible here also to