464 



SCIENCE. 



[Vol. V., No. 122. 



half of the curve is higher than the first, in the 

 ' 5 exaggeration ' the two halves (except in the 

 case of the colored people) are about equal. 

 The ' 10 exaggeration ' is thus rather a char- 

 acteristic of old age, while the ' 5 exaggeration ' 

 is used try old and young. 



There remains another peculiar irregularity 

 of the census figures which deserves special 

 treatment. It is the excess of the number of 

 persons at 21 over those at 20 years. This 

 excess is not of the same nature as the 

 ' 10 ' or ' 5 ' exaggeration, and is due, of course, 

 to political reasons ; 21 being the voting-age, 

 and 1880 the year of a hot presidential cam- 

 paign : accordingly this exaggeration ought to 

 occur in males alone. This is really the case. 

 In estimating the size of this excess, we encoun- 

 ter a difficulty. To compare the number at 21 

 with the number at 20, would probably be 

 comparing one exaggerated number with an- 

 other ; and, knowing that the number at 19 is 

 too small, we cannot make a fair comparison 

 with it. It is sufficient to notice, however, 

 that there are alwa} r s more males (and fewer 

 females) at 21 than at 19, and, when the ' 10 

 exaggeration ' at 20 is not large, more than at 

 20. Taking into consideration the excess at 



20, we have to declare the native male whites 

 ( the most reliable class in the former exagger- 

 ations) as the class that exaggerates most at 



21, — a conclusion quite natural, because they 

 are most apt to be benefited b} T such falsity of 

 returns. With regard to states, the inhabit- 

 ants of the extreme west ( Dakota, Wyoming, 

 etc.) would rank as the worst, the New-Eng- 

 land states as the best, under this head. 



Whether this exaggeration is increasing or 

 decreasing, is a question which unfortunately 

 can be only very partially answered. Previous 

 to 1880, the returns on age were given mainly 

 in groups of five years. In 1870, however, all 

 persons above 80 years of age were enumerated 

 by single years. This makes possible a com- 

 parison between the excess of the number at 

 90 over that at 89 in 1870 and in 1880. This 

 comparison is entirely in favor of the census 

 of 1880. In this decade the exaggeration at 

 this particular age (90 over 89 ) has fallen, for 

 the total population, from 104.6% to 65.7%. 

 As to sex, the male excess has fallen, from 

 87.1% to 36.7% ; the female, from 118.7% 

 to 90.3%. The colored people, too, have 

 decreased their excess very greatly, — from 

 1267% to 647%. Two other peculiarities in 

 the returns of 1870 may be noted : first, the dif- 

 ference between exaggerations of the sexes is 

 less, disappearing entirely in the colored race ; 

 second, the excess in the native whites is 



exceptionally high, being 155.3%, while in 

 1880 there is no excess at all, but a deficiency 

 of 4.8%. 



The observation of such facts as have been 

 here noticed, it is hoped, will shed light on 

 the characteristics of the natural bias in favor 

 of round numbers, as well as be a means of 

 suggesting modifications in the method of ques- 

 tioning which would obviate these misrepre- 

 sentations. It is just such irregularities that 

 detract from the value of the census figures 

 with regard to the calculation of the life-period, 

 and expectation of life, in the United States. 

 A more thorough comprehension of the ques- 

 tions treated above will doubtless be attaina- 

 ble from the census reports of 1890. 



Joseph J astro w. 



CHOLERA INOCULATION. 



A letter from Dr. J. Ferranof Tortosa (Catalonia) 

 to the French academy (Comptes rendus, ISTo. 15, 

 1885) contains some interesting assertions in regard 

 to cholera and the cholera bacillus. He finds that 

 cultures in bouillon at 37° C, carried on long enough 

 to just visibly change the fluid, will, in doses of from 

 two to four cubic centimetres, kill a guinea-pig. 



At the point of inoculation appears a hot and 

 painful tumor, which dries up and becomes detached, 

 leaving an ulcer behind, which heals without pus 

 formation or pain. The general symptoms are a 

 rapid rise of temperature, bringing on a lowering of 

 the physiological heat as taken in the rectum. 



If a drop of blood be taken from an animal thus 

 inoculated, and during life, and this drop be inocu- 

 lated in bouillon, kept at 37° C, in from twenty- 

 four to forty-eight hours a pure culture of spirilla 

 will be obtained. 



Microscopic examination of the serous effusion, 

 coming after a blow upon the inoculated side, shows 

 the following : — 



1°. Extraordinary number of globules, so much so 

 as to make one doubt the nature of what is being 

 observed. Many of the red-blood globules have pro- 

 jections, and possess a real movement due to the 

 striking of the microbes against these points. 2°. 

 Spirilla and commas, almost invisible by reason of 

 their rapid movements. 3°. Spherical cells full of 

 granulations, some of them containing a granulation 

 resembling a degenerated blood-cell. 4°. Lenticular 

 elements, varying from five to twenty millimetres in 

 size, and differing from the others described above. 



A series of cultures in gelatine preserves its viru- 

 lence, whilst a series in bouillon becomes attenuated 

 after a certain time. If a series of guinea-pigs be 

 inoculated with a quantity of the culture less than 

 sufficient to kill them, they become capable of resist- 

 ing doses which would before have been fatal, — a 

 result which the writer claims he has obtained. 



Effects of the microbe upon man. — The injection 



