36 



NATURE 



[May 14, 1 89 1 



(except that of the winter irregularity, 1889-90), I could 

 assign a fairly plausible explanation. For instance, 

 during the summer of 1886 I was under medical treat- 

 ment : in July of 1890 I was touring among the Swiss 

 mountains: while at the end of February 1887 I had 

 removed from a low-lying northern suburb, to a rather 

 higher southern one ; this change might with some 

 plausibility be considered as the possible disturbing 

 cause in the 1889 spring curve. 



Nevertheless, looking at the results as a whole, I was 

 not satisfied with the curves : it appeared to me as by no 

 means improbable that the monthly average was calcu- 

 lated on a rather too short period, thus allowing tem- 

 porary disturbing causes to manifest themselves unduly. 

 I therefore determined to try the effect of calculating 

 the averages on a two-monthly period, throwing into one 

 total January and February, March and April, May and 

 June, July and August, September and October, Nov- 

 ember and December respectively. On drawing the 

 curves corresponding to these averages (/^zV/^-lined 

 curve), I was delighted to find order and symmetry 

 completely regnant : all the aberrations have of course 

 disappeared, and order is supreme. This two-monthly 

 curve clearly shows a single maximum in winter, followed 

 by a fall to the minimum at midsummer, and then by a 

 rise to the winter maximum. 



It is evident that the curves for all five years are very 

 closely similar, though by no means identical ^ in nature ; 

 but I am especially anxious to point out the extraordinary 

 symmetry displayed by the curves on either side of a 

 inaximum or minimum point. For instance, the curves 

 for the following periods. 



1888 November-May 1889, 



1889 The whole year, 



1886 July-October, 

 ,, April-December, 

 ,, November- February 1887, 



are wonderfully symmetrical, in some cases even being 

 almost geometrically exact. 



What, however, may be the exact interpretation of 

 these curves I must leave it to those better acquainted 

 than myself with fphysiology to decide ; but it is worth 

 noting that these curves are exactly contrary to the 

 statement in Michael Foster's text-book, that the pulse 

 is said to rise in summer."^ 



The following is the table above referred to as showing 

 the net data for each month, and also (in brackets) the 

 number of readings below 60 included in each case : — 



' Seeing how very many causes must co-operate in producing the one 

 ■esultant— of pulsation-rate— it would be^very strange if tlie curves for 

 different years luere identical. 



^ I quote from the third edition. I 



