Agarella gracilis. 215 
is formed only at the end of the spore development, ‘‘in der Copula findet 
jedoch eine Kernverschmelzung nicht statt, vielmehr teilen sich die Kerne 
weiter bis die Zahl von 8 (monosporer Typ) oder 14 (disporer Typ) erreichtet 
ist [2, p. 205]. 
Now as to the formation of the “gametes” I have no new evidence to 
offer, but certain figures seen by me in the developing spores of Agarella 
gracilis seem to indicate another method of origin of the pansporoblast 
(disporous type). The presence of large and small cells, Auerbach’s macro- 
and micro- gametes, which copulate is certain (Pl. VIII. Fig. 5); but one can 
find also such unions as those depicted in Figs. 6 and 7, which would indicate 
further fusion of such copulee in pairs. 
A description of the figures seen in smears will be given first, and in this 
description certain interpretations will be given to the various stages. As 
these lead to a method of regarding the Myxosporidian pansporoblast which 
differs from that of recent authors, notably Auerbach and Georgevitch [6], the 
reasons for the interpretations here given will be discussed afterwards. 
According to Auerbach and others the macro- and micro- gametes are 
formed by division of macro- and micro- gametocytes, and the same is true for 
Agarella gracilis, but as there is nothing new here the first stages which concern 
us (Pl. VIII. Figs. 2-4) are the macro- and micro- gametes of Auerbach and 
others, while Fig. 5 represents the union of these two cells. Nothing resembling 
Mercier’s description of the formation of a synkaryon at this stage was seen. 
Now Pl. VIII. Figs. 6 and 7 represent stages which can be interpreted most 
naturally as the fusion of two such cells as Fig. 5, while Fig. 8 shows the 
resulting cell or syncytium with four nuclei derived from four different cells. 
Support is lent to this interpretation of the figures by the fact that the 
size of the developing pansporoblast (Fig. 9) is much nearer that of two united 
sporoblasts than it is to that of a single sporoblast. (Fig 5 shows an 
unusually large sporoblast.) From this point onwards the pansporoblast, 
as it is now called, proceeds as usual by repeated nuclear divisions to 
form the complex arrangement of two spores into which it is transformed 
finally. The fate of the two small nuclei during this process would 
be a point of some importance, but it is difficult to be certain in this 
matter, owing to the fact that all the nuclei become small before the final 
eighteen-nucleate stage, and hence it is difficult to pick out the original 
small pair. The appearances in Agarella gracilis indicate that the two small 
nuclei become the pansporoblast nuclei of Keysselitz [8], “ Restkerne” of 
Auerbach [2]. In the later stages they usually are peripheral in position 
and lie between the two sporoblasts. When eighteen nuclei have been 
formed great changes occur in the pansporoblast. The spore walls appear for 
