NATIONAL ORIGINS PROVISION OF IMMIGRATION LAW 5 
the statistics of mother tongue for the 1920 and the 1910 censuses; 
and, of course, that subdivides the statistics for each country in terms 
of the number of each mother tongue from that country, and gives 
us a picture not simply of the foreign-born themselves, which were 
also reported, but reflects really the immigration of a considerable 
period, the period in which their immigrant fathers came to this 
country. 
Senator Remp. To the extent that you can check those figures 
against the mother tongue, that is more certain than the 1890 census, 
is it not? 
Mr. Bocas. I think that without any question it is. 
Senator Rep. So that you have got the first of your four ele- 
ments settled in terms of post-war geography. You have got the 
second one subject to pre-war geography corrections? 
Mr. Bocas. Yes. 
Senator Regn, But checked by mother-tongue statistics? 
Mr. Bocas. Yes. 
Senator Reep. In getting your 1890 basis you do not have that 
check by mother tongue? 
Mr. Bocas. No; because our mother-tongue statistics begin in 1910. 
These two, may I add, account for one-third of the total quota. 
Senator Reep. To the extent of that third, you have a greater 
certainty under the national-origins basis, I take it, than you have 
ander 1890°¢ 
Mr. Boges. That is true. 
Senator Reep. How about the remainder? 
Mr. Bocas. Dividing between the two, the colonial stock and the 
“ grandchildren ” (as we speak of it) part of the post-colonial or 
immigrant stock: The Census Bureau has done a very great amount 
of work in trying to make the division between those two as precise 
as possible, and find that the ratio is about 2 to 1; in other words, 
about 45 per cent in the quotas get their distribution from the colonial 
Jk of 1790, and about 21.6 per cent from the grandchildren 
actor. 
Actually, whatever element of uncertainty there is in dividing 
between the two has very little effect. Accurate tests show that if 
there has been an error of 1,000,000 population in dividing between 
the colonial and the grandchildren factor (in the computation by 
the Bureau of the Census) which is rather difficult to suppose, the 
effect on the quotas would amount to less than 1,000 in the case of 
Great Britain, less than 400 in the quotas in the cases of Germany 
and the Irish Free State, and less than 50 in each of the rest. So 
that the division between those two, although it has been done with 
great care, really has not as much effect upon the quotas as would be 
supposed. , 
Senator Regn. Now, Mr. Boggs, turning to the 1890 method—that 
is, the basis of determining the quotas according to the foreign born, 
shown by the census of 1890—is that certain, definite, and accurate? 
Mr. Bogos. I take it that the figures as reported are accurate. We 
have no reason to question them. 
Senator Reep. The figures in the census itself? : 
Mr. Boggs. The ficures in the census itself: yes, sir.