THE COLLAR-CELLS OF HETEROCGLA. 19 
the surface with the cell in full profile, often seemed to show 
beads corresponding in number with the collar fibrils. On the 
whole, I believe that the radiation represents a condition exist- 
ing in life. Vosmaer (19) figures a ring near the base of the 
collar (“at the base ”’ in explanation of plate) in Spongilla, 
of which he promises a description ; it is not beaded. 
The substance which stains in this annular manner I shall 
call the iris, and the aperture in its centre the pupil. It is 
a natural suggestion that the iris is a contractile sphincter, 
aud the pupil the ingestive and egestive aperture of the cell. 
Some sections support the view that the thickening at the base 
of many flagella is cell-protoplasm projecting in an ameeboid 
cone through the pupil, the true flagellum running in the 
axis unthickened to the nucleus. 
Nucleus. 
There is nothing exact concerning the nucleus to be re- 
corded from the observations of living cells. I have above 
referred to the “empty vacuoles” of figs. 5 and 9; one of 
similar position is shown in fig. 3. If the identification be 
correct, these indicate (1) that the nucleus is distal in life, 
_ (2) that it moves in the protoplasm. In a drawing made in 
life from the same preparation as fig. 13, of cells with very 
active flagella, there is a large clear sphere in each cell which 
can scarcely be other than a nucleus. 
Preparations stained in bulk with borax-carmine show in 
the nuclei of collar-cells a well-defined chromatin reticulum 
surrounded by a stained nuclear membrane. In the wall of 
one chamber was a beautiful karyokinetic spindle; presumably 
the rather large cell in which it occurred was a collar-cell 
dividing in two. 
In 8. compressum hardened for one hour in 1 per cent. 
osmic acid, and stained carefully in bulk with hematoxylin, — 
the nuclei are almost always spherical; the radix of the 
flagellum can be recognised as a refractile thread passing from 
the nucleus to the pupil of the iris. The same series of sections, 
stained also on the slide with hematoxylin and extracted with 
