i io “ Turtle-Back ” Celts in District of Columbia. [February, 
The latter is the class to which I desire to call special attention. 
These lie scattered along various small ravines formed by the rain, 
mingling with pebbles and modern relics ad libitum. They are 
all made of characteristic yellowish or grayish-brown quartzite, a 
material used in every instance, which has not been found to 
occur in a single individual of the two preceding varities. In 
form they are true “ turtle-backs,” if the term is admissible. The 
variety of forms thus far discovered are represented by that unde- 
finable shape usually termed celts , and spear-heads. 
An examination of a celt gives the following measurements, in 
inches: length 4.6, width (greatest) 2.4, thickness 1.85. The 
anterior surface, or that side upon which we find the “ turtle- 
back ” elevation, rises to 1.38 above the average line of the cut¬ 
ting edge; while the posterior or opposite side rises to but .47 of 
an inch beyond the same line. The top of the greater elevation 
retains part of the natural worn surface of the cobble-stone from 
which the specimen was wrought, showing the implement to have 
been manufactured near the locality. The cutting edge though 
still sharp, is very undulating, owing to the removal of large 
flakes during manufacture. Secondary chippings, ending in a 
perfectly continuous edge, shows the relic to have been a com¬ 
plete specimen and not a core, as has been suggested upon the 
discovery of similar forms. Another reason is, that not a single 
arrow-head or other specimen has as yet been found, even upon 
the closest examination, although the flakes would serve to fur¬ 
nish materials for arrow-heads, which could measure, when 
finished, nearly two inches in length. Considering the nature of 
quartzite, such utilized flakes should be found, as no disintegra¬ 
tion or decomposition could obliterate their form so long as the 
bodies from which they have been flaked fail to present any 
changes upon their surface. 
The only spear-head found, thus far, presents the same peculiar 
irregularity, but is especially interesting on account of its great 
comparative thickness. The specimen is 2.5 inches in length 
from the projection on the base to the point, 1.8 of an inch broad, 
and 1.2 thick. The cutting edge is sharp, continuous all around, 
and slightly undulating, the latter being the result of flaking. 
Secondary chippings are visible at different points. The anterior 
lateral cutting edges converge at an angle of 90°, terminating in 
an extremely delicate and acute point. The “ turtle-back ” eleva- 
