NOVEMBER 26, 1903] 
NATURE 03 
the generative nuclei. But is such an_ interpretation 
possible? I think not. 
On harvesting the plants the grain was found to be badly 
shrivelled, the Michigan Bronze crosses only producing three 
grains, none of which germinated. From about three 
hundred grains of the Rivet and Red King crosses, two 
hundred and sixty plants were raised. The rust appeared 
on these as early as March 16, and by June 15 many plants 
were orange-coloured even on the highest leaves. On 
counting out the plot, 78 plants were found to be free from 
disease, 118 were slightly infected, and 64 were badly 
attacked. By June 29 the epidemic seemed to be at its 
height, and a second count showed that the number of 
disease-free plants was reduced to 64, while 195 were in- 
fected, for the most part badly.’ ‘ : 
These figures seem to be too close an approximation to 
the Mendelian ratio of 1: 3 to be a mere accident, especially 
when taken in conjunction with the results of the first 
generation. The susceptibility of wheat to the attacks of 
rust is therefore a definite Mendelian character. 
If further researches should show that this capacity for 
resisting the attacks of disease-producing fungi is in reality 
a tangible characteristic, the plant-breeder, at all events, 
will have definite lines to go upon in attempting to solve 
one of our most important agricultural problems, namely, 
that of producing disease-resisting strains. 
= R. H. BIrren. 
PROBABLE ERROR IN VITAL STATISTICS. 
PAPER on “ The Degree of Accuracy of Statistical 
Data,”’ by Mr. Carl C. Engberg, has been published 
by the University of Nebraska. ‘‘ This paper,’’ Mr. 
Engberg tells us, “is written as a protest against the un- 
necessary refinement of statistical computations as carried 
out by the biometricians of to-day.’ Mr. Engberg com- 
plains that the more ‘“‘ prominent biometricians have 
worked with five or six figures when they might have 
worked with three or four with equally good results. He 
illustrates this by comparing Prof. Pearson’s work on 
enteric fever, published in 1894, with a revision of it by 
himself using only three places of decimals. He considers 
that the one is as good as the other. He does not, how- 
ever, apply the test for relative goodness of fit of observ- 
ation to theory—Phil. Mag., July, 1900—but discards it 
without examining the analysis by which it is reached, on 
the basis of a paradox that he has not been able to' see 
through. He appears to dislike the test because if 16,000 
observations are distributed in the same proportions in n 
groups as 1000 observations the former distribution shows 
a lower probability for the fit than the latter, if the same 
curve be used in both cases. This, however, must be right. 
16,000 observations should give a result nearer a smooth 
curve than 1000. The percentage error has been discarded 
for years by trained biometricians; it was merely a 
temporary modus vivendi. 
As to the use of a greater or less number of decimal 
places, to those who work with mechanical calculators the 
number is practically indifferent, and to trained computators 
even a 7 or ro figure table of logarithms is hardly slower 
in use than a 4 or 5 figure table. But are the decimal 
places when reached worth having? Very often not, very 
often they are. Mr. Engberg seems quite incapable of 
distinguishing between the two classes of casés. The only 
means of testing is to consider the probable errors of the 
results. The theory of the probable errors of the constants 
of frequency curves was not given until 1898, and it was not 
possible to say in 1894 how many places of decimals were 
cr were not necessary. Mr. Engberg appears to think that 
because vital statistics are tabled to one or even four year 
units, it is impossible to ascertain the values of constants 
to two or three decimal places of those units. He states, 
vaguely, that “‘ the constants cannot be more accurate than 
the data upon which they are based.’’ He might well have 
asked the American actuaries for their views on this point ! 
Practical men do not work to 6 or 10 figures for the mere 
pleasure of it, and in the particular case cited by Mr. 
Engberg—*‘ Makeham’s c’'—we have a constant which 
1 One plant overlooked. 
NO. 1778, VOL. 69] 
has often to be raised to the fiftieth power! ‘Does Mr. 
Engberg believe that the mean age at death of a population 
of several millions, classified solely by year of age at death, 
cannot be found to less than the rough year which is the 
basis of the grouping ? 
Mr. Engberg says that his ‘‘ paper has not been written 
in a fault-finding spirit by a detractor of the new science 
of biometry, but by a teacher of the science.’’ It seems a 
pity that Mr. Engberg should not have studied either the 
history or literature of the science he is teaching, or, 
apparently, have received a training in mathematical 
statistics. In the former case he would have known that 
the method he suggests on p. 9 has been long in use (Yule, 
Proceedings of Royal Statistical Society, vol. Ix., part iv., 
1897), and the inner meaning of tables of powers for testing 
the last column of high moment tables would have been 
obvious to him. In the latter case he would have made a 
valid criticism of over-many decimal places by simply show- 
ing that they gave results beyond the probable error of the 
constant involved, or did not improve the goodness of fit 
as tested by a legitimate method. We are doubtful whether 
the growing practice of appointing teachers of biometry in 
the American universities without preliminary training is 
really helping the science forward. It leads, it is true, toa 
multitude of biometric papers, but very few of these are of 
permanent scientific value. 
UNIVERSITY AND EDUCATIONAL 
INTELLIGENCE. 
Oxrorp.—On Thursday, November 19, an election took 
place of great importance in relation to the advancement of 
the study of mental science in Oxford—that of Dr. W. 
MacDougall to the readership in psychology founded some 
years ago by Mr. Wiide. The first reader was Mr. Stout, 
the well-known editor of Mind, who last summer resigned 
the Oxford appointment for a professorship at St. Andrews. 
His successor will no doubt follow in Mr. Stout’s footsteps, 
but those who appreciate the value of the experimental 
method in psychology confidently anticipate that oppor- 
tunity will now be given for this study in Oxford. Dr. 
MacDougall has himself approached the subject from the 
experimental side, and is the author of very important re- 
searches on the physiology of the nervous system, among 
which those relating to the theory of colour vision may 
be taken as examples. As a member of the Cambridge 
Anthropological Expedition to Torres Straits and Borneo, he 
conducted experimental researches on the mental pro- 
cesses of savages, which afford substantial evidence of his 
power as an investigator. When to this it is added that 
he is no iess a master of the older methods than of the 
new, and is regarded as a man of power alike by physio- 
logists and by philosophers, the university may well be 
congratulated on so valuable an accession to its intellectual 
forces. Against any regret that may be felt that the new 
reader is not an Oxford man, the fact may be set off that 
one of the most successful and influential of American 
experimental psychologists—Prof. ‘Litchener—received his 
training in both sides of the subject at Oxfords 
CampripGe.—Dr. Hobson, F.R.S., has been appointed 
the first Stokes lecturer, and Dr. Baker, F.R.S., the first 
Cayley lecturer, in mathematics. 
An Isaac Newton Studentship of 2001. a year for three 
years, for research in astronomy and astronomical physics, 
will be vacant in the Lent term, 1904. Candidates must be 
bachelors of arts under twenty-five years of age. Applica- 
tions are to be sent to the Vice-Chancellor by January 26. 
The degree of D.Litt. is to be conferred honoris causa on 
Prof. Théophile Homolle, member of the Institute of 
France, director of the French School of Athens. 
Prof. Woodhead, Mr. J. E. Purvis, Dr. Tatham, Dr. 
Lane Notter, and Dr. R. D. Sweeting have been appointed 
examiners in State medicine for the diploma in public 
health. 
The scheme for the establishment of a geographical 
school and the institution of a special examination in geo- 
graphy and a diploma in geography will be submitted to 
Senate for adoption on December s. 
