W. R Macdonell 
205 
The weights, in the order in which they were taken, were 766*10, 766*80, 764'22, 
763-15, 768-95, 769-75, 766-70, 760-60, 764-25, and 764-30 grammes, the average 
being 765-48, and the difference between maximum and minimum weight being 
about 1^ per cent. This average, it will be observed, agrees pretty closely with 
the average obtained from the " cranes etalons," but as it shows that the density 
of seed in the glass was somewhat less, say a half per cent, less, than that in the 
skulls, some correction should be made in the capacities which I recorded of the 
31 skulls first ineasured. I have, as a matter of fact, reduced them one per cent., 
because I feel confident that, when working on these skulls, I subjected the 
measuring glass to a less thorough process of shaking and tapping than later 
on, when I came to make the ten experiments just described*. 
In order to ascertain whether the state of the weather affected the weight 
of the seed, I put aside a quantity weighing about 756 grau)mes, and weighed it 
on six different days in January last year ; two of the days were bright and dry, 
the others damp and cloudy. The variation in weight was less than half a gramme, 
and may therefore be considered negligible. 
I may add that the process of determining the capacity, whether by the glass 
measure or the balance, was very slow ; I never succeeded in taking more than 
three capacities in an hour. 
I am fortunately able to compare the result of my method and my personal 
equation with those of some of the other workers at University College. Some 
four or five years ago G. U. Yule measured 48 of the same crania. These crania 
give the following results : 
Yule Macdonell 
21c?s 15.31 1543 
27cJs 1321 1329 
In the averages of 21 and 27 cases our differences were 12 and 8 cm.^ 
respectively. Our average differences in reading were 20 cm.^ in the first and 
18 cm.' in the second series. I have also the measurements of 11 skulls made 
by A. Martin Leake. His mean value for the eleven was 1389 as against my 
1400, and our average difference was 11 cm.'. His average difference from G. U. 
Yule on these skulls was 8-6 cm.^ and the latter made the mean 1385. Yule 
* It is perhaps worth while pointing out that different results are obtained according as the vessel is 
filled full or only partly filled with seed, before the tapping and shaking process begins. For instance, * 
I made the following experiments with a 1000 cm.^ glass measure, 64 mm. in diameter : 
(1) The vessel was filled with seed, without shaking or tapping, and then tapped and shaken; 
when subsidence had apparently ceased, it was again filled up, and the packing went on again until it 
was finally filled up. The contents were found to weigh 759-60 grammes. 
(2) The packing process began when the vessel was three-fourths filled; after subsidence it 
was filled up, then tapped again, and finally filled up with tapping. Weight of contents 775-62 
grammes. 
(3) The vessel was two-thirds filled without shaking or tapjjing and then tapped ; filled to about 
seven-eighths full and tapped ; finally filled up. Weight of contents 776-98 grammes. 
(4) Similar to (.3), the stages being one-third full, one-half full, and full. Weight of contents 
778-52 grammes. 
