514 



SCIENCE 



[N. S. Vol. XXVI. No. 66S 



average, to ± 0.002 see. The average dis- 

 tance between the spark point and the paper 

 is about .5 mm. and the average angle of 

 deflection of the spark is less than half of 

 45°; therefore the average variation in the 

 spark is less than =t 0.001 sec. 



The chronoscope may be adapted for the 

 measurement of longer intervals, as in the 

 study of association, by two minor changes 

 which can be made in a minute. A small 

 weight is fastened on the top of the upper bob. 

 This makes the pendulum swing so slowly 

 that it takes three seconds to cover the arc 

 of the scale. A corresponding scale, gradu- 

 ated empirically in hundredths of a second, is 

 clamped over the regular scale. The accuracy 

 is nearly proportional to the speed of the 

 pendulum. 



Similarly, if there should be a demand for 

 finer reading's than those obtained by the 

 standard adjustment, an extra weight may be 

 placed on the lower bob that will cause the 

 pendulum to cover the arc of the scale, for 

 example, in one third of a second. If the 

 corresponding scale is graduated in thou- 

 sandths of a second each unit will occupy, on 

 the average, 1 mm. of space. The degree of 

 accuracy will be nearly proportional to the 

 speed, because the latent time of the spark is 

 negligible and the action is frictionless. 



Much of the value of a chronoscope lies in 

 its adaptation to the attachment of a variety 

 of accessories. The possession of the sound- 

 less make and break contacts for the stimulus 

 circuit makes it possible to connect all sorts 

 of electric stimulus apparatus, such as the 

 telephone receiver, the touch key, the tachisto- 

 scopes, etc. 



For regulating time-exposures, a movable 

 pendulum contact is attached to the front of 

 the base and adjusted, by reference to the scale, 

 for any desired length of exposure from a 

 hundredth of a second to three seconds. This 

 contact may be used either as a make or break 

 and the circuit may be completed either 

 through the make or the break of the stimulus 

 contacts. 



C. E. Seashore 



The Univebsitt of Iowa 



A VECTOR DIAGRAM 



Apropos of Carl Barus's interesting note in 

 Science of August 2, p. 149, it may not be 

 amiss to call attention to a representation that 

 I used in a communication to the March 

 meeting of the Chicago Section of the Amer- 

 ican Mathematical Society.' 



I represent a real point {x', y') in the plane 

 by a dot and call it a black point, while an 

 imaginary point {x' + ix", y' + iy") is repre- 

 sented by a blue point coincident with the 

 real point {x' + x", y' -)- y'') and joined to 

 {x' , y') by a real vector. Where no confusion 



1 



^y=V3[ 



is caused the real vector is drawn straight, but 

 otherwise it may be curved, it being under- 

 stood that the direction is determined by the 

 end points, rurthermore, if the vector moves 

 its end points describe a black curve and a 

 blue curve. Thus the line 



is represented by joining every point in the 

 black line 



2/ = 



to every point in the blue line 



2/ = l. 



In the accompanying diagram the " blue line " 

 is drawn heavy, the " red lines " broken. 



Ellery W. Davis 



QVOTATWNS 

 living on our CAPITAL 



The passion to beat our records in material 

 advancement tends to blind the thought to the 

 fact that we are rapidly consuming the very 

 fundamental resources on which the prosperity 

 of the country rests. Without doubt the 

 timber supply of the United States is disap- 

 pearing far more rapidly than any increment 

 of growth. The treatment of the soil in much 



' Bulletin of the American Mathematical So- 

 ciety, June, 1907, p. 436. 



