MUNSON: SPERMATOGENESIS OF PAPILIO. 
45 
and allowed to remain from fifteen to twenty minutes after which 
«/ 
it is transferred to 50 percent alcohol and carried gradually up to 
95 percent alcohol. It is then dehydrated and imbedded in paraffin 
in the usual way. I have used both the Toma and the Minot micro¬ 
tome, making the sections as thin as possible, usually from four to six y. 
Besides the corrosive acetic mixture which has been most satis¬ 
factory, I have, of course, used many others — picronitric and picro- 
sulphuric. Unusually clear asters and spindles can be obtained also, 
by first leaving the fresh testis in normal salt solution for fifteen minutes 
and then transferring to corrosive sublimate (pi. 16, figs. Ill, 112). 
Staining has been done almost exclusively on the slide after fixing 
to the slide by means of Myer’s albumen fixative. Corrosive sublimate 
material I have allowed to remain in a 50 percent alcoholic solution 
of iodine for about ten minutes either before or after staining. 
The following stains have been found useful, about in the order 
named: (1) Grenachers haematoxylin, (2) Delafield’s haematoxylin, 
(3) saffranin, (4) eosin, (5) acid fuchsin, (6) lithium carmine, (7) 
Biondi-Ehrlich’s triple mixture, (8) Haidenhain’s iron-haematoxy- 
lin, (9) orange G, (10) blue lumiere, (llj borax carmine, (12) Weigert’s 
picrocarmine, (13) Bisinark brown, (14) iodine green, (15) Congo red, 
(16) dahlia violet, and others. 
The larva and chrysalis have been studied entire without dissecting 
out the testis. By placing in the fixing fluid the fourth to the eighth 
abdominal segments, good fixation is secured. 
The Drawing. 
On comparing my drawings with those of Henking (’ 91 ), Paulmier 
(’ 99 ), Wilcox (’ 95 ), Bessels (’ 67 ), Meyer (’ 49 ), and some earlier writers 
on insect spermatogenesis, I notice that my figures appear diagram¬ 
matic. But a comparison of my drawings with the living cells of a 
freshly teased testis will prove to anyone that nothing can well be more 
diagrammatic than those cells in their normal living state. As all my 
figures with the exception of figures 1, 2, 3, 7, (pi. 12) were drawn with 
a camera of the latest and most perfect construction, I consider the 
regularity of my drawings as indicating perfect fixation. In fact, I 
know this to be the case, for I have studied the effect of my reagents by 
applying them to the living cell under the microscope. Many killing 
and hardening reagents ordinarily used, especially if allowed to act 
