(8) The two species of Lepidoptera examined have an equal pair of heterochromosomes.

FOOTNOTES:

[A] AUG. 20, 1906.--Since this paper was prepared, 19 other species of Coleoptera have been studied. Of these, 17 have an unequal pair of heterochromosomes in the spermatocytes. Six belong to the Chrysomelidae, making 14 of that family that have been examined. Representatives of 4 new families--Melandryidae, Lamiinae, Meloidae, Cerambycinae have been studied. In only two species--1 Elater and 1 Lampyrid--has the odd chromosome been found in place of the unequal pair. No species of Coleoptera has yet been examined in which one or the other of these two types of heterochromosomes does not occur in the spermatocytes. Of the 42 species of Coleoptera whose germ cells have been studied, 85.7 per cent are characterized by the presence of an unequal pair of heterochromosomes in the male germ cells, 14.3 per cent by the presence of an odd chromosome.

COMPARISON OF RESULTS IN DIFFERENT SPECIES OF COLEOPTERA.

In number of chromosomes there is great variation, the smallest number (16) having been found in _Odontota dorsalis_, and the largest (40) in _Silpha americana_. The difference in size is also very marked, as may be seen by comparing the spermatogonial plates in figures 3 and 58 with those shown in figures 94 and 141.

No other species of the Tenebrionidae has yet been secured, and all of the other beetles examined differ in a marked degree from _Tenebrio molitor_ in the growth stages of the spermatocytes. While in _Tenebrio_ the chromatin stains very dark throughout the growth stage, and the unequal pair can not be distinguished until the prophase of division ('05, plate VI, figs. 171-180), in most of the others there are very distinct synizesis and synapsis stages, following the last spermatogonial mitosis, then a spireme stage in which the condensed unequal pair of heterochromosomes or the odd chromosome is conspicuous in contrast with the pale spireme, whether the preparation is stained with iron-haematoxylin, gentian, or thionin. In _Tenebrio molitor_, the unequal pair behaved in every respect like the other bivalent chromosomes. In the other forms, though it behaves during the two maturation divisions like the symmetrical bivalents, it remains condensed during the growth period like the "accessory" of the Orthoptera, the odd chromosome, "_m_-chromosomes," and "idiochromosomes"

of the Hemiptera. In several cases the heterochromosomes of the Coleoptera are a.s.sociated with a plasmosome (figs. 22, 23, 63, 132, 158, 217), as is often true in other orders. This peculiar pair of unequal heterochromosomes varies considerably in size during the growth stage in some of the species studied, but changes very little in form, differing in this respect from the "accessory" in some of the Orthoptera (McClung, '02) and from the large idiochromosome in some of the Hemiptera (Wilson, '05).

The odd chromosome, so far as it has been studied, behaves precisely like the larger member of the unequal pair without its smaller mate (figs. 219, 220, 226, 233). In the growth stage it remains condensed and either spherical or sometimes flattened against the nuclear membrane (figs. 217, 225, 231). In the first maturation mitosis it is attached to one pole of the spindle, does not divide, but goes to one of the two second spermatocytes (figs. 233, 235). In the second spermatocyte it divides with the other chromosomes, giving two equal cla.s.ses of spermatids differing by the presence or absence of this odd chromosome.

All of the evidence at hand leads to the conclusion that in the Coleoptera, the univalent elements of all the pairs, equal and unequal, separate in the first spermatocyte mitosis and divide quant.i.tatively in the second. In this respect the behavior of the chromosomes in this order appears to be much more uniform than in the Orthoptera and Hemiptera.

COMPARISON OF THE COLEOPTERA WITH THE HEMIPTERA AND LEPIDOPTERA.

As has been seen above, the conditions in the Coleoptera, so far as the heterochromosomes are concerned, correspond very closely in final results with those in the Hemiptera heteroptera and the Orthoptera. In minor details these chromosomes are less peculiar in the Coleoptera than in either of the other orders. Even condensation during the growth stage is not universal, and synapsis of the heterochromosomes apparently occurs simultaneously with that of the ordinary chromosomes, instead of being delayed, as in many of the Hemiptera heteroptera.

_Aphrophora_ (Hemiptera h.o.m.optera) agrees with the _Anasa_ group of the Hemiptera heteroptera in having a pair of condensed _m_-chromosomes, in the growth stage, but this pair is already united in synapsis when first seen. It differs from _Anasa_, but agrees with _Banasa_ and _Archimerus_ in exhibiting a typical odd chromosome which goes to one pole without division in the first spermatocyte, and divides with the other chromosomes in the second spermatocyte. The odd chromosome in this species of Hemiptera, therefore, behaves like that in the Coleoptera and Orthoptera. The most interesting points in the results of this study of the germ cells of _Aphrophora_ is the discovery of two pairs of condensed chromosomes in certain phases of the growth stages of the oocytes. This has not been shown to be the case in any other species of Hemiptera, so far as I can ascertain. It is now evident that in the Heteroptera h.o.m.optera there are at least two distinct cla.s.ses as to behavior of chromosomes. In one cla.s.s we have the Aphids (Stevens, '05 and '06) and Phylloxera (Morgan, '06) in which no heterochromosomes have been found, while in the other cla.s.s are such forms as Aphrophora with both a pair of _m_-chromosomes and a typical odd heterochromosome.

The two species of Lepidoptera examined indicate that here we may have conditions comparable to those in _Nezara_--an equal pair of heterochromosomes whose only apparent peculiarity is their condensed form during the growth stage. Doubtless the results of other investigators will soon throw more light on the heterochromosomes of this order.

GENERAL DISCUSSION.

It will be seen from the foregoing that the results obtained in the study of the germ cells of _Tenebrio molitor_ have been confirmed in full for several species of Coleoptera, and in part for 19[B] different species belonging to 8[B] families. It has also been shown that a different type of Coleopteran spermatogenesis exists in at least 3 families, where an odd chromosome like that in the Orthoptera occurs in place of the unequal pair. In all of these insects the spermatozoa are distinctly dimorphic, forming two equal cla.s.ses, one of which either contains one smaller chromosome or lacks one chromosome.

The most difficult part of the work has been the determination of the somatic number of chromosomes in the male and female. In some cases suitable material has been lacking; in others, though material was abundant, no metaphases could be found in which the chromosomes were sufficiently separated to be counted with certainty. In three species (in addition to _Tenebrio molitor_) where the unequal pair is present, the female somatic cells have been shown to contain the same number of chromosomes as the spermatogonia, but an equal pair in place of the unequal pair of the male. In two new cases the male somatic number and size have been shown to be the same as in the spermatogonia. In one of the Elateridae, where the spermatogonial number is 19, the female somatic number is 20, and in _Aphrophora_ the numbers in male and female cells are respectively 23 and 24. No exception has been found to the rule established by previous work on the Coleoptera (Stevens, '05) and on the Hemiptera (Wilson, '05 and '06), that (1) in cases where an unequal pair is present in the male germ cells, it is also present in the male somatic cells, but is replaced in the female by an equal pair, each component being equal in volume to the larger member of the unequal pair, and (2) in cases where an odd chromosome occurs in the male, a pair of equal size are found in the female. It is therefore evident that an egg fertilized by a spermatozoon (1) containing the small member of an unequal pair or (2) lacking one chromosome, must develop into a male, while an egg fertilized by a spermatozoon containing the larger element of an unequal pair of heterochromosomes or the odd chromosome must produce a female.

Whether these heterochromosomes are to be regarded as s.e.x chromosomes in the sense that they both represent s.e.x characters and determine s.e.x, one can not decide without further evidence.

Comparison of the two types in Coleoptera, especially where, as in the Carabidae, both occur in one family, has suggested to me that here it is possible that the small chromosome represents not a degenerate female s.e.x chromosome, as suggested by Wilson, but some character or characters which are correlated with the s.e.x character in some species and not in others. a.s.suming this to be the case, a pair of small chromosomes might be subtracted from the unequal pair, leaving an odd chromosome. The two types would then be reduced to one. It may be possible to determine the validity of this suggestion for particular cases by observation or experiment.

Since the first of this series of papers was published, there have appeared three important papers by Prof. E. B. Wilson, bearing on the problem of s.e.x determination in insects. These papers are based on a study of many species of the Hemiptera heteroptera. These insects fall into two cla.s.ses--one in which a pair of "idiochromosomes," usually of different size, remain separate and divide quant.i.tatively in the first spermatocyte, conjugate and then separate in the second maturation mitosis; and another cla.s.s in which an odd chromosome--the "heterotropic" chromosome--divides in one of the maturation mitoses, but not in the other. Wilson regards the odd chromosome as the equivalent of the larger of the "idiochromosomes," its smaller mate having disappeared. In the somatic cells of the former cla.s.s he finds in the male the unequal pair, in the female an equal pair, the smaller chromosome being replaced by an equivalent of the larger "idiochromosome." In the latter cla.s.s the male somatic cells contain the odd number, the female somatic cells and oogonia an even number, the h.o.m.ologue of the odd chromosome of the male being present and giving to the female one more chromosome than are found in the male.

In his latest paper Wilson ('06) makes a variety of suggestions as to s.e.x determination. He shows that if the "idiochromosomes" and the heterotropic chromosome be regarded as s.e.x chromosomes in the double sense that they both bear s.e.x characters and determine s.e.x, the following scheme accounts for the observed facts in all cases where an unequal pair or an odd heterochromosome have been found:

Sperm. Egg.

{Large [Male] "idiochromosome"} I. {or } + Large [Female] s.e.x chromosome = a [Female]

{Odd chromosome. }

II. {Small [Female] "idiochromosome"} {or } + Large [Male] s.e.x chromosome = a [Male]

{No s.e.x chromosome }

Here we know that such a combination of gametes must occur to give the observed results, but we are not certain that we have a right to attribute the s.e.x characters to these particular chromosomes or in fact to any chromosomes. It seems, however, a reasonable a.s.sumption in accordance with the observed conditions. The scheme also a.s.sumes either selective fertilization or, what amounts to the same thing, infertility of gametic unions where like s.e.x chromosomes are present. It also a.s.sumes that the large female s.e.x chromosome is dominant in the presence of the male s.e.x chromosome, and that the male s.e.x chromosome is dominant in the presence of the small female s.e.x chromosome. Or, it might rather be said that these are not really a.s.sumptions, but inferences as to what must be true if the heterochromosomes are s.e.x chromosomes. This theory of s.e.x determination brings the facts observed in regard to the heterochromosomes under Castle's modification of Mendel's Law of Heredity ('99).

The question of dominance is a difficult one, especially in parthenogenetic eggs and eggs which are distinctly male or female before fertilization. It may be possible that the s.e.x character of the egg after maturation is always dominant in the fertilized egg, as appears to be the case in these insects (see scheme). Conditions external to the chromosomes may determine in certain cases, such as Dinophilus, which s.e.x character shall dominate in the growing oocyte, and maturation occur accordingly. It is evident that this reasoning would lead to the conclusion that s.e.x is or may be determined in the egg before fertilization, and that selective fertilization, or infertility of gametic unions containing like s.e.x characters, has to do, not with actual s.e.x determination, but with suitable distribution of the s.e.x characters to future generations. If both s.e.x characters are present in parthenogenetic eggs, as appears to be the case in aphids and phylloxera, dominance of one or the other must be determined by conditions external to the chromosomes, for we have both s.e.xes at different points in the same line of descent without either reduction or fertilization.

Wilson suggests as alternatives to the chromosome s.e.x determinant theory according to Mendel's Law, (1) that the heterochromosomes may merely transmit s.e.x characters, s.e.x being determined by protoplasmic conditions external to the chromosomes; (2) That the heterochromosomes may be s.e.x-determining factors only by virtue of difference in activity or amount of chromatin, the female s.e.x chromosome in the male being less active. The first of these alternatives is an attempt to cover such cases as _Dinophilus_, _Hydatina_, and _Phylloxera_ with large female and small male eggs. Here Morgan's ('06) suggestion as to degenerate males seems much to the point. The male s.e.x character, having become dominant in certain eggs at an early stage, may, from that time on, determine the kind of development. As to the second alternative, I see no reason for supposing that the small heterochromosome of a pair is in any different condition, as to activity, from the large one. The condensed condition may not mean inactivity, but some special form of activity. And, moreover, it has been shown that in certain stages of the development of the oocyte of one form, _Aphrophora quadrangularis_, there are pairs of condensed chromosomes corresponding to those of the spermatocyte, so that there would hardly seem to be any basis for Wilson's attempt to a.s.sociate the difference in development of male and female germ cells with activity or inactivity of chromosomes, as indicated by condensed or diffuse condition of the chromatin.

On the whole, the first theory, which brings the s.e.x determination question under Mendel's Law in a modified form, seems most in accordance with the facts, and makes one hopeful that in the near future it may be possible to formulate a general theory of s.e.x determination.

This work has been done in connection with a study of the problem of s.e.x determination, but, whatever may be the final decision on that question, it brings together a ma.s.s of evidence in favor of the belief in both morphological and physiological individuality of the chromosomes, as advocated by Boveri, Sutton, and Montgomery. It also gives the strongest kind of evidence that maternal and paternal h.o.m.ologues unite in synapsis and separate in maturation, leaving the ripe germ cells pure with regard to each pair of characters.

BRYN MAWR COLLEGE, _June 7, 1906_.

FOOTNOTES:

[B] AUG. 20, 1906.--36 species belonging to 12 families. See note, p.

49.

BIBLIOGRAPHY.

BOVERI, TH.

'02. Ueber mehrpolige Mitosen als Mittel zur a.n.a.lyse des Zellkerns. Verh. d. phys.-med. Ges. Wurzburg, N. F., vol.

35.

CASTLE, W. E.

'03. The heredity of s.e.x. Bull. Mus. Comp. Zool. Harvard College, vol. 40, no. 4.

MCCLUNG, C. E.

'99. A peculiar nuclear element in the male reproductive cells of insects. Zool. Bull., vol. 2.

'00. The spermatocyte divisions of the Acridiidae. Kans.

Univ. Quart., vol. 9, no. 1.