APPENDIX 



358^ 



of these results is quite unintelligible unless it be pathological, as in the case described by 

 Hertwig. It may be recalled that Flemming discovered typical tetrads in the spermato- 

 cytes of the salamander, but regarded them as "anomalies" (cf. p. 192), and the later 

 studies of Meves and McGregor render it probable that his interpretation was correct (cf. 

 preceding note). These highly interesting observations probably open the way to a future 

 better understanding of the tetrads; but for the present they leave the whole subject of 

 reduction in a state of confusion worse confounded. 



17. P. 194, line 12. Meves (I.e.) apparently disproves the accounts both of Flemming 

 and of vom Rath, and shows that the reduced number appears in but two cell-generations, 

 as in Ascaris. 



1 8. P. 197. Calkins (Bull. Torrey Dot. Club, XXIV. 3, '97) has discovered typical tetrads 

 in the spore- formation of ferns (Adiantum, Pteris). The tetrads arise, as in insects and 

 copepods, by one longitudinal and one transverse division of a primary chromatin-rod; and 

 the facts indicate further, that the actual reduction is effected by the second cell-division. 

 It is an interesting fact that the tetrads vary in mode of origin, some of them forming rings, 

 others arising from completely split rods, which afterwards divide transversely. 



19. P. 199, 1. 29. In a highly interesting paper (Jahrb. wiss. Bot., XXIX., '96) Klebahn 

 has shown that in the diatoms (Rhopalodia) reduction is effected in the course of two 

 rapidly succeeding nuclear divisions occurring before conjugation. After union of the conju- 

 gating cells the nucleus of each divides twice (maturation-divisions) to form four nuclei, as 

 in many Infusoria. Each cell now divides into a pair of gametes, each containing two 

 nuclei; of the latter one disappears (cf. the " corpuscules de rebut"), while the other forms 

 the germ- nucleus. Each gamete now conjugates with one of the other pair, thus forming 

 two zygotes (auxospores), the two germ-nuclei in each fusing to form a cleavage-nucleus. 

 The number of "chromosomes" in the two maturation-divisions is four; that occurring in 

 the ordinary division of " vegetative cells " is certainly larger at least six and probably 

 eight. 



20. P. 200, 1. 34. Studies made by Mr. F. C. Paulmier, in the Columbia laboratory, on 

 the spermatogenesis of Anasa and some other Hemiptera give results different from those 

 of Wilcox. His preparations clearly show that the tetrads arise from chromatin-rods that 

 undergo a longitudinal split, followed by a transverse division, essentially in the same 

 manner as in Gryllotalpa or in the copepods. 



21. P. 226, 1. 20. Several recent workers seem somewhat disposed to abandon or to 

 impose certain limits to the law of genetic continuity in case of the centrosome; and they 

 have given more or less definite accounts of its free formation. Among such writers may be 

 mentioned Auerbach (/. Z.,XXX., '96), Watase (Science, V. no, Feb. 5, '97), Child (I.e.}, 

 Mead (I.e.'), and Lillie (I.e.}. In an earlier paper (/. /I/., XI., '96), I myself was inclined to 

 adopt such a position, though with some reservations; and my more recent observations 

 on sea-urchins (see note to p. 228) may seem to support it. I am nevertheless of Heiden- 

 hain's opinion (Morph. Arb., VII. I, '97, p. 202), that in this difficult field of research the 

 positive evidence is entitled to the foremost place, and that only the most exhaustive and 

 convincing research will justify a negative conclusion. The present controversy regarding 

 the centrosome recalls the long debate, following the promulgation of the cell-theory, on 

 the question of free formation or genetic continuity in case of the nucleus. The outcome 

 of that debate may teach us caution in this case. 



22. P. 228, 1. 3. The structure of the centrosome in echinoderms remains an open 

 question. Kostanecki's recent observations (Anat., Hefte VII. 2, '96) on Echinus agree 

 with those of Hill on Sphcer echinus, the sperm-centrosome being a simple granule which 

 divides into two halves, and these are shown in the plates as far as the late prophase of 

 cleavage. In Arbacia I have found essentially similar facts (Science, V. 114, March 5, '97), 

 and have traced the centrosomes through the anaphases and telophases, during which 

 they divide and give rise to daughter amphiasters, as in 1'halasscma, Fhysa, or Chietopterus. 



