This description does not give much detail nor do the drawings show the stages clearly. As far as given, the formation of the nebenkern is the same as in _Gryllus_; but in _Gryllus_ the axial filament is not formed from the nebenkern, as can be plainly seen from fig. 17. Erlanger (’96), in a short paper, discusses the use of the term “nebenkern,” and suggests limiting it as is done in this paper. He opposes St. George’s opinion, that the nebenkern comes from the cytomicrosomes. In _Blatta_ the cytomicrosomes are preserved during the whole process of mitosis, and have no connection with the spindle fibers, but during the telophase they collect in reduced numbers around the daughter nuclei. In 1897 he called the collection of granules around the centrosome, the centrodeutoplasm. He considers them to be identical with St. George’s cytomicrosomes and the archoplasm (or attraction sphere) of other writers. In order to harmonize results, he suggests that, since the centrosome sometimes wanders around the nucleus, the centrodeutoplasm (or sphere) may unite with the spindle remains in some cases to form one body, as shown by the descriptions of Henking, Henneguy, Meves, and others. He later (’97, 2) describes the so-called “sphere,” and distinguishes between it and the true nebenkern. I would strongly commend his excellent discussion of literature and his careful comparison of the results of investigators. He has shown clearly that the nebenkern comes from the spindle remains. Calkins (’96) finds that the nebenkern comes from the spindle fibers and is useless in the cell. But _Lumbricus_ is peculiar in having the nebenkern simply disintegrate, for, in many cases, he admits that the nebenkern has an important function. I do not have access to Henneguy’s or Bolles-Lee’s or Toyama’s works, yet I should judge from Erlanger’s and Meves’s criticism that all of these have the nebenkern originate from the spindle remains, and Henneguy describes it as having a “fibrillar appearance,” and Bolles-Lee as “fibrillar structure.” Accordingly, I think that each of these has discovered the correct origin of the structure, and I do not doubt that there is, at least in the first two, a more or less direct change from the spindle remains to the nebenkern. Paulmier (’99) finds that in _Anasa_ the nebenkern comes from the yolk mass and remains of the spindle fibers. A part of this mass separates off, while the whole is still in a confused condition, and forms the acrosome. The nebenkern forms the tail sheath, while the acrosome forms the point to the head.