404 
its turn closed and rounded off, and to 
this finally the two whiplike flagellz are 
attached. These are first drawn out 
from a glass rod into straight threads 
about the thickness of a fine needle, 
and then are curved by passing the 
glass threads through the flame sev- 
eral times in different directions. Now 
the glass stem retained at the lower 
end of the outside cup for a handle, is 
cut off short to serve as a concealed sup- 
port for attachment to the final mount. 
Models for the other fifteen individuals 
composing the colony are now con- 
structed in the same way and arranged 
on the mount in their proper places — 
and the model is complete. 
The model of the radiolarian further 
illustrates methods of glass-modeling. 
From a solid glass rod one-quarter inch 
in diameter, nine smaller rods are fash- 
ioned each about one-eighth inch in 
diameter and about four inches long. 
These are somewhat bent and attached 
at equal distances to the rim of a per- 
forated cup previously blown to repre- 
sent the skeleton of the central capsule. 
In this way the principal framework of 
the skeleton is prepared. In order to 
give it greater stability during the work, 
slender glass supports are welded tem- 
porarily to the lower ends of the rods. 
Then beginning at the central capsule 
small glass rods are welded horizontally 
to connect the larger elements of the 
skeleton, until finally the whole network 
is completed as shown in the figure of 
the finished model. 
Many protozoa are beset with count- 
less hairs or cilia, and in representing 
these the welding of such closely set, fine 
structures on the models involves great 
difficulty. Even on models represent- 
THE AMERICAN MUSEUM JOURNAL 
ing great magnification these ciliafare 
often so fine that a little careless’ smanipu- 
lation with the blast lamp will cause 
whole areas of the heated cilia to col- 
lapse, mowed down like grass before the 
mower’s scythe. 
Although molten glass may i brought 
into hundreds of different shapes, never- 
theless the methods of blowing are 
always practically the same. It is the 
setting together of the separate parts 
however which requires great care and 
alertness, for when once the parts are 
wrongly joined they can be corrected 
only with difficulty or not at all, and it 
is usually necessary to reconstruct the 
entire piece from the beginning. 
In many cases, where several parts 
are welded together, the finished struc- 
ture must be thoroughly annealed. 
This is best done by greatly diminishing 
the air pressure in the blast lamp, when 
the glass parts are rotated steadily for 
some time in the smoke and flame of 
the weakened jet. This is necessary 
because in working the glass for a long 
time, alternately thinner and _ thicker 
places will occur. These produce an 
uneven tension and the glass will break 
if it is not carefully annealed. 
As mentioned above the methods of 
glass blowing are very simple. Only 
skill in the worker is necessary to pro- 
duce the most diverse shapes from the 
molten material. In order to attain 
this skill, years of training for hand and 
eye are necessary. The calling of the 
glass-blower so to speak is an inheri- 
tance from antiquity. The sons grow 
up to the father’s trade and devote 
themselves from early youth to the ac- 
quisition of the all-important feeling 
and skill. 
