20 



SCIENCE 



[N. S. Vol. L. No. 1279 



paragraphs under the two species I show 

 maxima named in 1775 and fusiformis in 1804. 



I wish very cordially to thank Professor 

 Cockerell for his kindness in calling these 

 errors to my attention and giving me the op- 

 portunity to correct them. 



Dr. Ellis L. Michael questions a statement 

 in the same paper (page 139) in which I say: 

 " The solitary individuals (of Thalia demo- 

 cratica) lie at a considerable depth during 

 winter, spring and early summer, coming to 

 the surface with the aggregated zooids in the 

 fall." He writes that the records of the 

 Scripps Institution show "the almost com- 

 plete restriction of both generations to the 

 months of Jime and July. I have gone 

 through our list of deep water collections 

 again, and find that the statement made in 

 my (published) report to the effect that, when 

 all depths are considered, the species is still 

 almost entirely restricted to the months of 

 Jime and July, stands as given." 



My statement quoted above was somewhat 

 inaccurate. Salpa (Thalia) democratica has 

 been found at the surface every month in the 

 year, but in ItTorth Atlantic waters it is most 

 abundant at the surface from July to Septem- 

 ber. When not at the surface the animals 

 must be in deeper water. A more accurate 

 statement than the one quoted would be that 

 both solitary and aggregated forms of Salpa 

 (Thalia) democratica are less frequent at the 

 surface during the colder months, becoming 

 more abundant as summer advances, and be- 

 ing most abundant in the late summer and 

 early fall. The conditions off the California 

 coast seem a bit exceptional, the time of 

 maximum frequency of this species at the sur- 

 face of the ocean being about a month earlier 

 than in North Atlantic waters, and the species 

 being less frequent in the winter, spring and 

 fall than in many regions. Dr. Michael's re- 

 port of its abundance in June and July and 

 its scarcity at other times, reminds one of 

 Agassiz's reference to the sudden appearance 

 of this species off the ISTew England coast and 

 its equally sudden disappearance.^ In few, if 

 3 ' ' Three Cruises of the Blake, ' ' Bull. Mus. 

 Comp. Zool. Soward Univ., Vol. 14, 1888, p. 190. 



any, other localities have so full records of 

 distribution of pelagic organisms been made, 

 as off La Jolla, and it may be that similar 

 complete records for this species for other 

 localities would show somewhat closer agree- 

 ment with the records of the Scripps Insti- 

 tution. 



Maynard M. Metcalf 

 The Orchabd Laboratory, 

 Oberlin, Ohio, 

 June 12, 1919 



"WORKING UP" IN A SWING 



A CHILD sitting or standing in a swing can 

 " work up " until he is swinging through a 

 considerable distance. How is it possible for 

 him, without touching his feet to the ground, 

 to increase the extent of his swinging? As 

 I do not recall ever seeing any discussion of 

 this matter, the following note may not be out 

 of place. 



What the child does appears to be this: 

 Near the end of an excursion he shifts his 

 position so that he is on the whole farther 

 from the axis of rotation [limb of tree, or 

 other support], and when he is near the middle 

 of his path he brings himself back again to- 

 ward the axis. Now a shift of matter either 

 away from the axis of rotation or toward it 

 changes the moment of inertia about that 

 axis, and therefore tends to change the an- 

 gular velocity. In fact, unless a large torque 

 is acting, a sudden shift must necessarily 

 change the angular velocity. If the shift is 

 made at a time when the angular velocity is 

 small the change in angular velocity is small, 

 but if the shift is made at a time when the an- 

 gular velocity is large the change in the an- 

 gular velocity may be considerable. Thus by 

 moving toward the axis when near the middle 

 of his path the child increases his velocity, 

 whereas by moving away when near the end 

 of the path he produces little change in his 

 velocity. 



This action may be imitated by a pendulum. 

 Instead of keeping the length of the pendulum 

 constant, the upper end of the suspending cord 

 is passed over a hook and is held by a hand. 

 The pendulum is set swinging with a small 



