818 
ROSS GRANVILLE HARRISON 
Still another feature of importance was shown in the prepara- 
tion under consideration. In the previous observations upon the 
growing filaments, it had been impossible to find absolutely fixed 
points from which to measure their increase in length, so that there 
remained the possibility that the extension of the fibers was not 
due to the active movement of the enlarged end but rather to the 
passive shifting of the cells from which the fibers were seen to arise. 
While such an objection would not hold in cases like the one just 
described, where the fiber is visible throughout its entire length 
from the cell of origin to the amoeboid end, such instances are 
not very common, and it is desirable to have other means of 
determining which structure is moving. The specimen under 
consideration has demonstrated that in good preparations where 
the clot is firm, the red blood corpuscles, some of which are nearly 
always to be found mixed in the lymph, can be used for this pur- 
pose. One particular case (text fig. 2) will serve to illustrate the 
point clearly. A short stout fiber (nf) was observed emerging 
from a mass of cells, its cell of origin not being visible. Near the 
end of the fiber (npl) was a group of five red blood corpuscles 
(1-5) arranged characteristically, and at the point of emergence 
of the fiber was another single corpuscle (6). In the vicinity of 
the end of the fiber were two large separate cells (cti, and ct2) 
derived from the implanted tissue. The relative position of the 
structures just mentioned is shown in fig. A. The end of the fiber 
was then only fairly active. An hour and fifty minutes later 
(text fig. 2 B) it had progressed only 25m, as measured by its 
relation to corpuscles 4 and 5 . The loose cells (cti and ctz) , however, 
had moved considerably, but in a direction approximately at 
right angles to the direction taken by the fiber, showing that they 
could not have been concerned in the movement of the latter. 
At the same time these cells had changed their shape materially, 
as is shown especially in cell ct2, this change in shape being undoubt- 
edly due to the activity of the hyaline ectoplasm. Six hours and 
twenty-five minutes after the first observation was made (text 
fig. 2 C) the distance through which the end had moved was 100^, 
i.e.] at the rate of 15.6/x per hour. This rate is slow as compared 
with that observed in other cases, though it is considerably faster 
