133 
A box or bowl filled with clean, dry sand will be found very useful to 
support the specimens in while the cement is drying. The broken pieces 
should be so placed that the force of gravity will hold them in contact. 
Plaster of Paris will be found necessary for some repairs and plastic 
wood may be used for filling in pieces of woodwork which are deeply 
scarred or from which small fragments are missing. 
RESTORATION 
Restoration implies the making of duplicates of missing parts of speci- 
mens. This has to be done with much caution and self-restraint. The 
intention is not to deceive, but to indicate the appearance of the specimen 
when it was complete. Sometimes the best method is to make a copy of 
the entire specimen and to exhibit it side by side with the incomplete 
original. The facts of the case are then clearly explained on the label 
accompanying the specimen in the exhibition case. Full details should be 
entered in the catalogue if any restoration work is done, to prevent future 
students from assuming that the restored specimen is in the condition in 
which it was collected. 
NUMBERING 
Catalogue numbers should be on the specimen itself whenever this is 
at all possible and, preferably, in an inconspicuous position. In selecting 
a place for the number it is advisable to bear in mind that it may at some 
time be necessary to photograph the specimen from various angles, and 
numbers should, therefore, be kept well away from important details of 
construction. A great saving of time and trouble is effected by adopting 
a standard system of numbering similar articles. Thus all baskets might 
bear the number on the bottom, all spoons underneath the tip of the bowl, 
all masks just under the mouth at the back of the mask, and so on. In 
this way one knows at once where to look for a number on most specimens. 
As a rule India ink shows up well, but when the object is black or 
nearly so, white water-colour paint is the best medium. When numbers 
have to be put on specimens that have a high polish or a vitreous surface 
there may be difficulty in making them distinct, as the ink has a tendency 
to concentrate in a number of droplets with dry spaces between them. 
This may be overcome by covering a sufficiently large area on the specimen 
with celluloid solution. When dry it will be found to provide a ground 
with an excellent “tooth” which takes the numbers well. When the India 
ink is, in turn, dry, another coat of celluloid is flowed lightly and quickly 
over it, and the number is thus locked between two celluloid layers. Entries 
in the catalogue should also be in India ink; if desired, a thinner grade 
than that used on the specimens, for many ordinary writing inks fade after 
a number of years, whereas every effort should be made to have the cata- 
logues as permanent as possible. 
In numbering large specimens, such as big boxes, canoes, totem poles, 
and so on, the number should be quite large but in an inconspicuous posi- 
tion. Few things are more annoying than having to search for half an 
hour for figures a quarter of an inch high on a 40-foot totem pole. 
