Page:Science vol. 5.djvu/497

 I .ivna b, ]S6S.)

��SCIENCE.

��461

��baci pivsented its broailside to the laiincb. As the boat approached bow on, it corresponded to a target somewbere about six feet square, preaentiDg a convex suiTace to the impinging sound-wave. Even in this case a feeble echo was i>erceiveii wheu the boat was sit a con- siderable distance (estimated to be nearly one- quai-lei- of a tnile). That any echo should have been perceived at all undei' such cii'cura- stanees, was a surprise. Thesound was heard only by the cloaeat attention, but in the case of latter vessels the effects were very distinct and striking.

ExperimentB were made which demonstrated the Tact that the speaking-trumpet attached to the gun was of material asaisuince iu giving direction to the souud-impulse, and in iuten- sifyiug the audible efiect.

Mr. Delia Torre claims that a a team- whistle or siren, combiued with a projecting apparatus like u speaking-trunipet, will prove as efllcient as the gun.

During ilie experiments on the Patapaco River, a curious rambling effect, like the roll- ing of thunder, was often observed, which con- tinued for some seconds. A similar sound was idao noticed, as an echo from a well-wooded shoru; but the ert'ect alluded to above could not have been due in any way to the land, as the sound commenced iiniuediateiy upon the firing of the gun, whereas the shore was dis- tant Jit least a mile or a mile and a half.

The sound was probably due to the presence of ripples on the surface of the water, as the effect was muoli less marked when the surface was smooth. Such a sound might prove a disturbing element of importance in a rough sea, but would l.urdly be sufficient to prevent the detection of an echo (Vora a large icebei^. Had shots lieen fired |>eriodically fiom the Ijow of the City of Berhn, it can haidly be doubtetl that the presence of an obstacle ahead would hare Viecn discovered in time to prevent the collision that actually occurred.

A[.EXA>I>ER GKAfUSI BkLL.

��Shobti.t after the issue of the present cen- sus reports, attention was called to the pecul- iar fact that very many more persons were recoi-ded as being just '20 or just .'lO years old than were as being 19 or 49. It is easy to see that there ought to be more persons living at any one year of life than at the ne.\t, — more at 7 than at 8 years of nfic. more at IH

��or 49 than at SO or .')ll. Of all the infanta less than a mouth old at the prescut moment, quite a large number will die before completing their first year; many of those then surviving will die before the end of their second year; and so on, there being fewer left iu each year than in the preceding year.

But all this is true only when certain con- ditions are satisSeil. The growth of tbe pop- ulation, it is assumed, is by natural increase alone, or nearly so. The number of foreign- born inhabitants, for instance, between the ages IU and 15, will be smaller (that of native Americans, of course, very much larger) than the number between the ages 20 and iJo, because so very many of the immigrants are, on arrival, between 20 and 35 years old. So, too. a war. or an eptdemio which is parlicnlarly fatal to persons between certain ages, might be the cause of an exception to the general rule, at least until the generation so affected had died out.

But the effect of any such circumstances on the census figures which are here dealt with, may, without hesitation, be r^arde<i as insig- nificant. The preponderance of the number of persons at the ages containing rouml num- bers, over the number at the age immediately preceding (this being rather an ' odd * number) , must lie ascribed to an entirely differcut kind of influence.

Before going farther, it is necessary lo ap- preciate how euormous the attraction towards i-ound numbers really is. Very uatnrally this attraction is gieatest towards the ages contain- ing multiples of 10, for then the numbers are ' roundest.' Subtract the number of persons recorded as jeara old fiora the number re- corded as 10, and express this excess in per- centage of the number at 9 years. Do this for the excess of the number at 20 over the num- ber at 19, of 30 over that at 29; and so on, the last being the excess of the number at 90 over the number at 89 years of age. The average of the 9 pereenlagcs thus obtained is what I will call the average * 10 exaggera- tion,' anyone of the |>ercentage8 of which it is composed being spoken of as the ■ 10 exag- geration ' at 20 or 30 oi' 60 years, as the case may be. This average for the total population of the United SUtes is 71 > %; and the several percentages of which it is the average vary from 9..") % to 126 %. This means, tbatinai«ad of finding fewer |»ersons recorded at any such - round ' age as 20. 30. etc., than at the age immediately preceding (19. 29). you would find, on the whole, nearly IJ times as many. You might find only 1 ,'a (an i-scess of 9.5 %)

�� �