150 DE. COBBOLD ON THE EYE OF THE COD-FISH. 
occurrence, and I have the satisfaction of knowing that they were 
in the first instance witnessed by other microscopical observers. 
The bacillar prolongations continue to extend themselves ; an outer 
capsule of the twin-cone (represented at d), which normally 
binds the halves together, very soon gives way, and the divisions 
next exhibit the appearance seen at e. In some cases the half- 
cones do not entirely lose their original truncated figure, and (as 
at f) the bacillar filaments, with their investing sacs, are pro- 
longed downwards and outwards in the form of a cylinder. Most 
commonly, I might almost say invariably, the ends of the prolon- 
gations curve outwards, each in the form of a hook, as shown in 
all the three last figures alluded to, and still more significantly in 
the illustration marked h, where the capsule of the twin-cone and 
the sac of the bacillar filaments have altogether disappeared, 
each half-cone having at the same time undergone a marked change 
of form. This gradual disappearance of the saccular extension of 
the twin-cone capsule around the filament may be seen taking 
place whilst the cones are under examination (as on the right of 
the cone marked /) ; and now also the two halves frequently sepa- 
rate (as at 7 and g), portions of the capsule sometimes remaining 
partially separated. Occasionally some of these phenomena occur 
independently, the half-cones remaining united, the central line 
of separation only becoming uneven or sinuous (#) ; whilst at other 
times the distortion is so peculiar, that the combined halves re- 
semble a plant-stomate with an interspace between them (7). In 
all cases the bacillar filaments are liable to break up at different 
places, but they are most commonly detached at their points of 
connexion with the twin-cone. Whilst watching the cone marked 
z, the decussating appendages here figured floated off, along with 
the granular particles, which were carried past it in a continuous 
stream. In one instance only have I observed that great exten- 
sion of the filament which I take to be the complete unfolding 
and development of this peculiar appendage ; and in this case, only 
the right half of the cone came under notice (m). After the sepa- 
ration of the filaments the half-cones undergo all manner of changes 
of form, and they ultimately split up longitudinally and disintegrate. 
Thus, the halves marked m and o have a regular oval figure, but 
the granular contents are seen to be enclosed in a separate enve- 
lope of a pear-shaped form. At p and qg they have assumed a 
rounded outline, and at 7 and s have broken up lengthwise, leaving 
in the latter a hollow centre, owing probably to the escape of the 
less dense albuminous molecular matter which normally occupies 
