SUMMARY,1897. MONTHLY WEATHER REVIEW. 671 attributing a uniform average velocity to all winds. These TABLEB.-Fmpncy of thundaP.eiolme mnd awmw during 1897. resultants are also presented graphically on Chart I, but - I i;; Number of 8 Total for Frequmcy should be studied in connection with both the lower isobare stations. +I 1881. per statton. e of Charts I and IV and the upper isobars of Chart V. The - LI - - relation between the resultant winds thus computed from twa ti i -c1 i observations per day, without regard to velocities and those ,: !L! computed from twenty-four hourly observations, taking full 8 E account of the velocities, can be estimated by a comparison -Pi -f -4 -3 between Tables V and VI, pages 644 and 6-45 of the Summary Alabama ............... 6.1 1% 46 8.8 415 0 9.2 0.00 48 for 1894. Arizona ............... 11.4 2&3 80 0.6 rn 1 10.7 0.08 I Arkansas.. ............. 6.2 180 45 2.9 672 0 14.9 0.00 42 -0- California .............. 16.8 896 115 8.4 a98 2 2.6 0.02 86 Colorado ............... 10.4 860 66 4. n 917 8 14.1 0.12 m FREQUEN- OF THUNDERSTORMS. Connecticut ............ 0.6 12 1I 0.8 884 18 14.9 0.87 45 Delaware.. ............. 0.2 5 4 1.2 (14 6 16.0 1.m 48 The successive MONTHLYWEATHER REVIEWS have given for District of Columbia.. 0.01 n. a 8 0.5 81 0 15.5 0. 00 41 each day and each State the number of thunderstorms re- Florida ................. 5.9 148 40 4.7 972 0 24.8 0. 00 44 Georgia.. .............. b.8 146 45 8.2 454 0 10.1 0.00 49 ported by both voluntary and regular observers; Tables VI Idaho. .................. 8.6 815 80 7.2 rn a 9.7 0.27 do Illinois ................. 5.5 188 0 1.7 1,878 17.9 0.80 58 and VI1 give the annual summary.of these monthly tables. Indiana.. ............... 8.4 85 45 1.9 800 !! 18.8 0.10 60 In order to ascertain the relative frequency of thunderstorme Indian Territory.. ..... 8.1 78 6 15.6 &3 i 18.6 0.00 86 Iowa ................... 5.5 198 90 1.5 918 44 10.1 0.49 46 for the whole country exhaustively, it would be necessary ta Kansas ................. 8.1 909 65 8.1 748 I 11.5 0.43 8B Kentucky ............. 8.8 95 40 2.4 Bot 0 15.1 0.00 m have at least one special thunderstorm observer for every Louisiana .............. 4.1 108 45 2.8 860 0 19.1 0. 00 41 20 miles in distance, or every 400 square miles of area. Maine .................. 8.6 w 16 5.9 101 e&i 6.8 6.67 I Maryland ............ 1.1 28 &l0.9 628 18 17.6 0.40 e The corresponding number for the respective States is given Massachusetts ........ 0.8 20 20' 1.0 25a I 12.6 8.15 61 Michigan ............... 5.6 140 80 1.8 739 103 9.1 1.38 57 in the third column of the accompanying Table B. In the Minnesota.. ............ $4 210 &l8.5 Bw) 101 11.6 1.88 ta absence of such a system of stations, it is proper to divide Mississi pi ............. 4.7 118 40 2.8 578 1 14.4 0.25 48 Misrourr.. .............. 6.5 162 80 2.0 1,812 12 a.6 0.16 44 the number of storms reported by the number of reporting Montana ............... 14.4 880 85 IO. 8 103 iw 6.6 8.00 b Nebraska ............ 7.0 190 90 2,1 718 28 7.9 0.81 44 stations in order to deduce the average number per station Nevada.. ............... 11.2 Wl 86 8.0 m 17 8.6 0.49 4 per annum. The results of this division are given in the New Hmpsblre ....... 0.9 22 15' 1.5 174 89 11.6 6.98 65 New Jersey ............ 0.8 30 45 0.4 7x3 14 17.1 0.81 48 eighth column of Table B, which shows that the greatest fie- Mew Mexico. ........... 12.1 908 80 10.1 286 0 9.5 0.00 48 New York .............. 4.7 11R io 1.7 717 86 10.2 1.58 m quencies per station per annum were: South Carolina, 24.9 ; North Carolina... ...... 6.1 128 60 8. 2 i.aw 0 21.0 0.00 4 Florida, 24.3 ; Missouri and Tennessee, 22.6 ; North Carolina, North Dakota .......... 7.6 185 49 4.6 245 m 6.1 5.m 47 Ohio ................... 4.0 Irn I25 08 1,680 79 18.4 0.88 65 21.0. The smallest frequencies were: California, 2.6 ; Wash- Oklahoma .............. 8.0 98 a0 4.9 161 6 8.0 0.80 40 Oregon ................ 9.6 858 45 5.8 190 4 4. !a 0. OB w ington, 3.9 ; Oregon, 4.2. Pennaylvania .......... 4.6 115 70 1.6 961 19 18.7 0.87 69 The product of the observed number of thunderstorms by Hhode Island.. ......... 0.1 2 5 0.4 a# 2 6.6 0.40 48 South Carolina.. ....... 8.4 I 80 8.8 748 1 24.9 0.08 I1 the reduction factors given in column 5 will give the approxi- South Dakota .......... 7.6 190 45 4.2 296 67 6.6 1.49 62 Tennessee ...... 4.6 115 40 a. 9 904 8 22.6 0. OB 41 mate total number of thunderstorms for the whole area of each Texas .................. 27.4 Is5 io 9.1 Lffl 0 8.1 0.00 43 State. Utah ................... 210 80 7. n Yixl 1 9.9 0.09 I Vermont ............... 25 I8 2.1 178 47 14.8 a. 91 68 There were no very severe tornadoes during the year, the Virginia ................ 152 85 4.8 489 0 14.0 0.00 46 Washin on 175 45 8.9 175 a6 8.9 0.80 60 one causing the destruction of a portion of the town of West v&inii1: 1: 1: 1::: 68 80 1.9 817 1 10.6 0.08 60 Chandler, Okla., on March 30, being the most notable. The Wiswnrin ............ 182 65 2.4 611 105 11.2 1.91 64 Wyoming. .............. 245 16 :6.4 141 5 9.4 0.88 62 year as a whole was remarkably free from violent local I storms. *The values for Connecticut New Ham hire and Massachusetts reduced from p- last year on account of disconhnuance of%e dnblication of a number of reports FREQUENCY OF AURORAS. from those States. -0- Tables VI11 and IX give a summary of the detailed tables of auroral frequency in the respective MONTHLYWEATHER SUNSHINE AND CLEAR SKY. REVIEWS. The annual numbers are also collected in Table . The succeeRive MONTHLY WEATHERREVIEWS have pre- €3. In the absence of more precise knowledge it is assumed sented in Table XI the percentages of sunshiiie a.8 recorded that the number of observers reporting all auroras is the same by either photographic or thermometric self-registers, as also as that of those reporting all thunderstorms, and is as given in Table I, the personal observations and estimates of the by the estimates published in the fourth column of Table B ; average cloudiness from sunrise to sunset. The corre- the number is, of course, decidedly less than the number of sponding chapters in the text have called attention to the those who report rainfall and temperature. systematic differences between the instrumental and the per- The total number of auroras reported divided by the num- sonal records. These differences are doubtless in part due ber of observing stations for any State gives the relative fre- to instrumental and personal peculiarities, such as arise in quency per station, as shown in the 9th column of Table B, every kind of exact work ; but in addition to these we must which number is comparable, with similar ratios for other consider the fact that the photographic and thermometric parts of the world, provided that the aurora is so low down registers give the dwatdon of certain limiting values of actinic in the atmosphere as not to be obscured by a cloudy eky. On and thermal effects respectively, whereas the personal obser- the other hand, if the auroral light emanates from a region vations give the percentage of area of clear sky. There is no far above the clouds, then a further correction for cloudiness simple relation between these three kinds of data and instead is needed. The average annual cloudiness at 8 p. m., seventy- of combining the records indiscriminately we should fir& fifth meridian time, is given in the tenth column of Table B, investigate the reasons for these differences. for regular Weather Bureau stations, but a correction for The differences (instrumental minus personal), as given in cloudiness has not been applied in the present case, as the detail in the tables published from month to month, are Editor believes that we have 110 certain proof of the extreme collected together in the accompanying Tables C and D for altitude of the aurora, while there are many reasons for be- the photographic and thermometric stations, respectively. lieving that the light emanates from the cloud region itself. A cursory examination of these tables shows that there is The States that reported the greatest frequency of auroras an annual periodicity by reason of which the differences are per station were: New Ha.mpshire, 5.93; Maine, 5.67; North larger in the summer thaii in the winter months.