may be traced to antecedent conditions that preexist in the germ-cells before their union. In most of the species the spermatozoa are predetermined as male-producing and female-producing forms, equal in number, that differ visibly in the constitution of the nuclei. The differences between the two classes do not appear in the mature spermatozoa but are plainly apparent in the process of their formation. In some cases the female-producing spermatozoa contain one more chromosome than the male-producing ones, in others both classes have the same number of chromosomes but one of them is much smaller in the male-producing class. These initial differences in the spermatozoa lead to corresponding differences in the nuclei of the two sexes, the cells of females either containing one more chromosome than those of males or showing a greater quantity of chromatin in the greater size of one of the chromosomes. The sex of the individual may, therefore, be recognized in these cases by simple inspection of the dividing cells.
Although these visible differences are of wide occurrence in these insects they are not always present, for in one of the genera male-producing and female-producing spermatozoa can not be distinguished by the eye and the nuclei of the two sexes have the same appearance. This case is, however, connected by almost insensible gradations with those in which the differences are plainly apparent, and it is hardly possible to doubt that sex-production conforms to the same type throughout the series. It is, therefore, not improbable that two classes of spermatozoa, predetermined as male-producing and female-producing forms, may exist in animals generally, even though they are not, as a rule, visibly distinguishable.
It is not yet known whether this also applies to the eggs before fertilization. In these insects there is no visible indication of such a predetermination, but several eases are known in which the eggs are of two sizes before fertilization, the larger ones producing females and the smaller ones males. It is, therefore, possible that in animals generally, both eggs and spermatozoa may be predetermined as male-producing and female-producing before their union. In any case these observations bring a strong support to the view, which has rapidly gained ground in recent years, that sex is predetermined at least as early as the fertilized egg, though they do not exclude the possibility that in some cases sex may be affected by conditions acting upon the embryo subsequent to fertilization. It does not yet clearly appear how these new results can be applied to an explanation of sex-production in parthenogenesis, in hermaphrodites and in such cases as that of the bee where all the fertilized eggs are of the same sex. A new basis has, however, been gained for the investigation of these questions, and also for an interpretation of sex-production in accordance with the Mendelian principles of heredity, the probability of which has been urged by Castle, Bateson and other writers.
TWO GERMAN DIRECTORIES
There appeared in November the fifteenth annual edition of 'Minerva' and a little earlier a German 'Wer Ist's.' The former of these works is well known to scholars throughout the world. It is an invaluable address book containing the names and chairs of the professors in the universities and other institutions of higher education of all countries, the officers of libraries, museums, academies, etc., and much information in regard to the organization of these institutions. The editing is a model of careful exactness; it requires some search to find a mis-spelled name among the 40,000 in the book. The work is brought out with unusual promptness. It appeared early in November and contains changes up to October. Each year a portrait of an eminent scholar is given as a frontispiece. This year Dr. Sophus Müller, director of the National Mu-
