III .
' On the Lunar Atmospheric Tide at Melbourne .
" By Dr. G. NEUMAYEIt , late Director of the Flagstaff Observatory , Mem. Acad. Leop .
Communicated by Lieut.-Gen .
SABINE , President .
Received April 10 , 1867 .
Anxious to assist the development of so interesting a branch of knowledge on the connexion of forces in nature as the influence our satellite exerts upon the earth 's atmosphere , I had made it a point to include investigations , tending to facilitate studies in this direction , in the plan of discussion of the observations made at the Flagstaff Observatory about to be published .
Fully aware that a geographical position , such as that of Melbourne ( 37§ 48 ' 45t south lat. and 9h 391 " 538 east long .
) , affords but very few chances for arriving forthwith at a result which might be regarded as final , I thought it nevertheless of the highest importance to decide how 2s 2 489 far and to what an extent such small oscillations as those in question , and which for lower latitudes have already been proved to exist , would make themselves manifest , in spite of the great atmospheric disturbances of higher latitudes .
The volume of discussions above referred to contains consequently the results of the reduction and classification of upwards of 43,500 hourly observations on pressure of air , registered during the period from the 1st of March 1858 to the 28th of February 1863 ; and in publishing these results I was chiefly guided by the conviction that it would hardly be compatible with the scope of such a work to enter upon a full discussion of the phenomena connected with the lunar influence on the barometer ; while a complete reduction and classification would make the observations apt to be taken up by any one interested in this matter for the purpose of being subjected to a rigorous examination and discussion .
While engaged upon this task , I could not fail , however , to be struck by some very interesting facts which , though they are far from being reducible to definite laws , may serve to furnish some connecting links with respect to atmospheric tides , and to give evidence as to the possibility of proving their existence even in as high a latitude as that of Melbourne .
A successful attempt at a complete solution of the problem may only be hoped for when a larger number of discussions on barometrical observations , collected at ectropical stations , will be at our command .
Prior to entering upon the task proposed , it appears desirable to give a few particulars , requisite for a full understanding of the subjoined results .
The geographical position of the Flagstaff Observatory was already mentioned , and it remains only to be added that the standard barometer was one of Newman 's construction , 0'400 inch in diameter , its cistern being 12017 feet above the mean level of the sea .
A few facts respecting the oceanic tides gleaned from the Sailing Directions for Port Phillip , ' by Capt. Ferguson ( 1861 ) , may also find a place here .
High water at full and change .
and fall .
and fall .
hm Spring .
Neap .
hm ft. ft. On the beach at Pt. Lonsdale ... ... ... . .
11 25 74 In the midchannel between Pt. Lonsdale and Pt. Nepean ... ... ... ... ... ... ... ... 1 50 At the Lightship , West channel ... ... ... .
2 10 43 At the East end of South channel ... ... . .
2 25 43 At the Bird rock , Geelong ... ... ... ... . .
2 30 3-5 2'5 At Pt. Gellibrand and mouth of River Yarra 30 4-5 25 There is no necessity for entering more fully into a description of the method employed in freeing the barometrical observations from the regular diurnal fluctuation and arranging the remainders b-b according to lunar time , inasmuch as this method is quite identical with the one employed by all who have directed their attention to this subject , as General Sabine , Professor Kreil , and others .
This much may be stated , however , that the reading of the barometer ( 6 ) , after being reduced to 32 ?
, was invariably increased by 1 inch , in order that it may always exceed the mean pressure for the respective hour ( b ) , thereby avoiding negative results .
This has certainly the disadvantage of not exhibiting at a glance the excess or defect of atmospheric pressure at any time ; on the other hand , there is no doubt that a mistake with regard to the algebraic sign of the remainders is not likely to occur .
In the subsequent Tables it was made a rule to reduce the values b-b to their mean value for a month , year , or whatever other period of time they may refer to .
The remainders b-b were derived , in the manner just pointed out , for every month throughout the period of five years for which the observations were continued .
Then the means for every month and hour were taken , thus obtaining normal values for the several months ; a general mean for every month formed the basis to which those normal values were referred .
The subjoined Table shows the result of this proceeding .
The values of the above Table have been thrown into curves , and Plate I. shows the results .
The actual mean values are indicated by dots ; a fulldrawn curve is made to pass through and between them in such a manner as to eliminate the greater irregularities .
Some of those irregularities are so large as to cause the respective dots to be disconnected with the series to which they belong , and it became therefore necessary to indicate this connexion by slight dotted lines .
On glancing over this series of curves we cannot fail to observe a great regularity , pointing at some cause common to all ; and as the remainders b-b have been arranged according to the moon 's hour-angle , we may justly look to the moon as the primary cause .
But it is nevertheless true that those curves apparently point to some other influence , most likely due to the combined action of the sun and moon .
The monthly curves for the several years of observation have also been drawn , though we refiain from adding the results here ; and the fact that they correspond in the main points with those shown on Plate I. , seems to justify our attaching particular weight to the evidence of the moon 's influence on our atmosphere , as conveyed to our minds by the above Tables .
But prior to entering fully upon the various points bearing on the question at issue , we need to form of the monthly results quarterly and semiannual groups .
If we call March , April , May the first , June , July , August the second , September , October , November the third , and December , January , February the fourth quarter , we obtain the quarterly means inserted in the following Table .
It was furthermore considered serviceable to the purpose to group together those quarters in which the epochs of solstices and equinoxes respectively occur , under the collective names " solstitial and equinoctial quarters .
" The semiannual periods comprise , as usually , the months from April to September , and those from October to March .
The mean of all the various hourly values represents the mean lunar-diurnal variation in pressure of air for the year .
1st part.--From the -superior to the inferior passage .
( 1858-63 ) .
~ont ; ~s .
h !
1 h , h 4h 5h 6b 7h h h9h ' II h Months. .
5h6 .
h. 9 .
7 9h .
in .
in .
in .
in .
in .
in .
in .
in .
in .
in .
in .
in .
April ... ... +o'o0210 -0o'oo00290 o-000150 -ooo64 --oo648 +0o00130 +000030 +o'ooIz4 ?
+000152 -0'00092 +0-00076 +000036 -May .
+0'00085 o+000673 --o0o005 +0'00233 -0-00051 -0'00221 +0o00219 +0'00031 -0'00377 -ooo00011 +0'00551 +000097 June ... ... +0'00508 0'00168 -o'00052 -0'00296 -o00432 -o'oo8o8 -000592 -0'oo812 -0'0o696 -o0oo376 -0o00264 +o000224 July ... ... +0'002180'00320 -oo-00536 -0'0059g -o'oo4i2 --oooi6g +0-00036 +0070 +o'oo522 +o'oo7 +0007124 -'0-00014 BAugust.-..ooo7-13 +o'oo437 00035 --o'oo00o -o-ooo37 -o'oo03 -2577 August 0007 1+000437 -0-00203 -0-00097 +000225'+ 353 +000137 +-00035-000141 -000347 000371 -000277 September +0'00474 214 00 +o'ooozo 0oo00138 +0-00498 -+0'00552 --o0oo52 +o00ooo2 +0-00042 -0'00342 +000050 -000230 October ... -o-00403 --000025 -00497 -0-00595 -oo0025 -000047 +-0o016 0-00117 +000303 -0-00059 +000139 +0o-o00265 November b -0-000676 0-o0o612 + --oozo --o'ooI4 -000406 +0-00098 +000282 -0-'00028 +000380 { 000728 +o000176 .
+000706 December o+0'00377 +0000257 +0-00457 -+o 'oo20 +00381 +0013I +0'00251 --0004.53 -00oo207 --0-01083 +0'00049 +0o00321 January -oo00178 -ooo568 -ooo3io --o000o68 +ooo +000324 -oo0026 +o'00ooo6 +000162 -0oo00644 --o0oo0032 +000152 February -0'0o465 +o00oo 0073 -0ooo00005 -oo00045 +000193 -0-00033 -0oo0145 -0 o00227 --000225 +o'00057 --000375 -0-00253 March ... .0-a00402 -0-00430 -0 o00290 +0'00024 +000124 +000156 +000516 +000966 000724 i+000536 -0'00108 --OO.008 2nd part.-Fromr the inferior to the superior passage .
( 1858-63 .
) ifMonths .
211 .
311. .
h411 .
75 " .
X68h i1 2Oh .
j 20 h1 221 .
23h .
in .
in .
in .
in .
i in .
in .
in .
in .
in .
in .
in .
April ... ... j +0'00162 +o'ooIO6+0 00'ooi38 +-0'0230 +0'00094 +0'00252 +000096 -000082 -00340 -0-00296 -0'00444 +00oo30 May --0'00033 +0'00357 +0'00311 +'00Ioo --0'00329 --o00o331 -0-00443 --o00oo28 --o00441 -000767 -0-o00103 +0-00939 June ... ... o'oo36 +000222+00078 -o'oo +00248 --oo00102 -o+000218 +oooi68 o+000164 +000136 +0oo00732 +0-00844 July ... ... i0-000072j -00364 --o00134ooI34 +o0oo 4.I +00o'ocoi8 --0'00044 --ooo162 -0-o00300 -000278 +0-00074 +0'00522 +000oo284 Augusut --.0002o 5 -0-00371 -000209 +0ooo00009 --00137 -0ooI6 --0'00377 +o'ooi0029 +000509 +0'00383 -0 -00063 -+0'00035 Septemnber +o oooo8 +0'00058 +0-00190 -o'oooi2 +'000o02 +00oooo8 -0-00284 -0-00782 -ooo00502 -00290 --ooo0029 --000540 iOctober ... i -0oo 79 --o'ooo8r +o0oo243 +o'ooi08 +0'00177 -000063 '+000249 -0-00007 -000127 +0'o01033 +000057 --o00603c3 November +0 00720 +0-00412 +-o00284 +ooooo68 +000500oo +000250 -00034.o -o-ooi4o --o'ooz88 -0-o00910 --o01032 --ooIo0101 December -0-0oo07o -000705 -o'0oo00 -0-o00341 +000143 +o'oo289 -0-o00109 -0-oooo00029 +000oo97 +0011o4-3 January. .
ooo00082 +000399 +0'00440 +0o00240 +oooi76 -000408 --00oo136 o+000242 +0'00358 -0-00ooo392 000116 -000394 February -00-o'oo4I1 -0ooo47 -o ' ooz27 -000ooI +-0002o75 +o00363 +o000473 +000o473 +000245 +000oo265 +o000225 +0-00097 March ... ... -00oo 226 +o0oooi6 +ooo86 --o000oo38 -0-00326 -0-ooo6o +0o00014 -0'00332 --o000292 +0-00028 -0'00214 -0o00388 1st part .
From the superior to the inferior passage .
( 1858-63 .
) 2h .
3h1 4h .
in .
i. i. in .
in .
nn .
I. Quarter -o0o00355 -0oo00055 -0-001848 -oocz000 I +o00o85 .
0oo004724 40o000952 --o002635 --0'003275 i-o'2o6 T , .
-0-o'0020o 7 -00 -0-04 -0002290 -o00oo043 -0-o000530 IV .
, , -0-00o870 -o00008I0 +0'000456 +0-000576 +oo002423 Equinoc , ) ot-co-o226 --ooooo66 -002o 00 -oo000563 -0-000205 Solstitial ) 010 00o 927 +0'0000oooo7 --0'089 -0-00I349 +O-Cooi8i Apr. to Sept , +0ooo368o 0+o0oo47o -0-001727 --O'-0460 -0-000317 Ot .
to M -ar .
-O-0029II -0'002174 0004244 -0000414 +0'000 332 Yer ... ... ... i+00039i -0-000346 -0-00o569 -0-00093 +0000ooooi4 51 .
6h .
71 .
81 .
9h .
in .
n. in .
i in .
i. +0o000z29 +0'002552 +0'003739 +c-ooi665 +o0o00445 -o0oz2075 -000oo395 --oo00202 I-0oIo048 -0-000015 +0-002010 +'ooo0997 +0-000303 +o-0024I7 1+o-oo0 090 +0o-00436 , -0-000417 -0'002234 -0-000917 -0'005584 -+00010841 +00-I 744 +-o001990 +0-0020Iz +0-001237 -0-o000319 1-o0ooogo6 -0-002z27 i-0-000982 -0-002799 ---000270 -0-000537 -0-000750 -0oo000830 1-0-000750 +o-ooo049 +0o'ooI4i2 +o-ooo652 +o'ooI896 i-0-000774 -+0000396 +o0ooo444 --o0000oooo42 000oo539 -00oo0756 1 Oh .
I I1 .
in .
in .
+0-00173z +00oooo72 +0-000298 +0-00022 I -000o42I 7 +0'002470 --0002524 +0000ooo7 6 +0o002944 +0-00129o -ooo00112 +-0'000248 +O'OOI277 -0'000273 +0-000592 +0-001849 +0-00094 +-oo000795 2nd part .
From the inferior to the superior passage .
( 1858-63 .
) h I. 13hI I511 I6h 17h 1.h 19h20 .
2I 2211 .
23h .
11 h. , -I6-- .
, __ in .
in in .
in .
in .
in .
in .
in .
in .
in .
in .
in .
I. Quarter -0'000322 +o-00oo599 +o'ooi785 +0oo00645 -o-oo868 1-0-00046I -o'ooII8 1-0-002315 -o003575 -0003448 -o2oo535 +0o00oo27 II , , -0-000468 --00oo708 l+O-oOI465 o+ooo05 o-0000432 -oooio -o'ooIo68 --000ooooo0008 -OOI3I +001979 +0'003972 0-o003879 IIIL , , +o0OOI830 0o'oo z297 +0o'oo239o 1+o0000790 +o0ooz563 +oo.oo65o -o00ooI250 -00oo3097 -0-003057 --0000557 -o'00418o 3 -0.003583 T , .
i -o'ooIi57 --oo001254 -oo002664 -0-001570 +O-OOOI49 -o0-00304 +-oooI583 +0'003329 1+ o0oo629 -0-000537 +o0oo4o09 J+o0oo543o Equinoc- .
-+o'ooo723 f-O'00147 +o'oo002o2 -+oooo687 o 0000317 +o-oooo0064 -o'oO2zo -0-002737-0003347 oo02033 --0'003390 -o'ooi686 Solstitial .
1 -0-o00ooo 1-0'001781 -0-000599 -000ooo0782z +oooz9I -0-001162 +o-000258 +o0-00661 +0o-001474 +o-00072z +o-004041 +0-004655 Apr. to Sept. --oo000007 +0+o-00003 +o-001797 9+0000367 -o 0-000023 --0'000630 -001oo587 -o0ooI093 -0-001480 -0-00I267 4+0'000590 +0o004620 Oct. to Mar. -ooooo44 -o-ooo206 -0o'oo30I -0-000424 o-0669 -000043 o672 +0000876 -0-000354 -ooooo8 +8 -0-0006I4 Year ... .
-0 ... .0ooooI9 -0'0o0090 +o +oco754 -0000o72 o000ooo329 j-0-oc0024 -0000451 -0-000512 -0o-092I -oo00063I 4+0000344 4+0-002009 ... ... 1I 005.._ .
7 ... 0 ... ... 00054 ... .05..0005 c The curves derived from the results of this Table are shown on Plate XI .
, with the exception of the semiannual and annual curves which may be studied on Plate X. Glancing at the various curves thus resulting , we are first struck by the great conformity of some of them , whilst others present irregularities apparently quite irreconcilable with what we feel inclined to adopt as the law .
There is , however , in all cases manifested a progressive change , evidently depending on the moon 's hour-angle in the first instance , calling for a rigorous examination .
The semiannual curves of the lunar-diurnal variation of atmospheric pressure may be taken as representing the principal types of the various monthly curves .
During the sun 's absence from the hemisphere ( in our case , when the sun 's declination is north ) , from April to September , the lunar variation reaches its maximum at about 23h 1 511 , or 45"1 prior to the moon 's upper transit , its minimum value occurring at 19h and a secondary one at 2 " , with a range of 0'00653 inch .
The curves for the single months appertaining to this semiannual period exhibit , generally speaking , the same characteristics , though somewhat irregular , and showing , in some instances , deviations of considerable extent ; so , for instance , the curves for August and September .
The summer semiannual curve ( while the sun 's declination is south ) exhibits an essentially different character , there being no strongly expressed maximum noticeable , whilst a decided minimum occurs at 01 30 " or 30 ; " past the moon 's upper passage , the maximal pressure taking place at 6 " , and a secondary one between 18 " and 19 ' .
The amplitude of oscillation amounts to 0'00432 inch .
But in this period of the year we notice a great difference in the lunar-diurnal variation of the barometer , when we examine the single months somewhat more closely ; thus , for instance , the curve for the month of December shows such characteristics as to cause it to be more like the curves for the winter period , and , on the other hand , we perceive that the curve for the month of November is exactly of the opposite character as that for December .
The remaining four months show more or less irregularity , and make a greater or smaller approach towards the general type for the class under consideration .
Although there is undoubtedly in all of these cases strong evidence of an influence of the moon on our atmosphere , I could not rest satisfied , considering that this evidence is seemingly of a somewhat conflicting nature .
As already explained , the monthly values have , for the purpose of further inquiry , been combined into quarters , and the results for these quarters were again united in mean values , arranging the two quarters in which epochs of solstice occur , and the remaining two , comprising the equinoxes , respectively , in two groups .
Thus we obtain six monthly mean values of the lunar-diurnal variation for the " solstitial and the equinoctial quarters .
" I was prompted to adopt this course because of the great similarity of the curves for December and June in the one case , and of March and September in the other , though in by far less a degree .
Tris similarity may best 4.:I 4 ... ~~ 4 .
IIi i\ i 'I14 .
, , I I. .
I V+ -i ; ---~-~-~~~i~-..YI Ii iI A/ j"~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ I ----rr rt--V-4.----rQ7*.44..4 4 - ?
-i~ ~~ ~~~~~I.p./ _~ ~ 1..1 II tI ~~~~~~~-------j / I j-r`~~~~~~~~~~~~~~~~~~~~~/ K 4..------- ?
... .-----(.-.J\ ; &gt; 7..I &amp; ... 7 .
-.-Lj r 4 ... 1h if..1*---ii it ?
' ?
:~ ~ ~~~~~~4..4- ?
1 t ?
-C ?
-m- ?
/j~ ~~~~~~~~~ .
/ : 4 .
/ II 4 .
4.4 N ... ... .~ L-{------i 4 , h S. 4f41 -i 41C--ii !
i`--J..4 ; 3t-----~~~~~~~~~~~~~ i , LIj -e t ii --r----------i~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~i4 Pp I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~4 , L__ia~~~~~~~ .
____~~~ ~ 4--44f j--4~ .
4+ .
i / , / , '-X i , i ~ -ai -~ I~-----I I. I ?
* ' i I &lt; +.C --- .
- .
-_ , I " - !
i / !
JIIi1i/ I I/ i ; :u / i !
i\ i il / , .
I/ ' 7 7pp r 12 3 h./ -4h -l 1 16 ; h/ i pqh Siq L7 ' ?
1 212h ' 23h , '(__ I : Jq I-'|---.'---'}----.-----.-JIQarte _ ... .-^^ ---_="..__ ... ... _ FQizacrter ---i---.-------3----WOu ; ^-_ i----.--_ _ i--.'_.eaune &gt; i i- !
1_ -iW -SI -- .
-----CeW part ( 01 m-eh h ' Z eh ; ' 3h 4h 57v h 7h .
g g7 " , h ni ib 12'. .
II+ _et ?
A--_ C ?
pper Tra-si &amp; .
be judged by the mean values for the respective two months , representing , as they do , in both cases a distinct oscillation with an amplitude of 0"'01722 for the solstices , and of 0"'01041 for the equinoxes .
I may be allowed to refrain from adding here these mean values , suffice it to refer to the respective curves at the bottom of Plate II .
For December and June we observe the maximium to occur at 2311 , the minimum at or shortly after 81 , while for September and March the maximum in the lunar-diurnal variation of pressure of air takes place at 7 " and the minimum at 19 " .
The mean for the quarters ( each embracing six months ) show the same characteristics , though by far less in extent , the amplitude for the solstitial quarters being 0"'007454 , and that for the equinoctial quarters 0 " 006334 .
There is another fact which requires to be pointed out , in order to throw further light upon the character of these oscillations ; namely , that they seem to bear a great resemblance for both hemispheres during the same semiannual period , if we are permitted to arrive at this conclusion by referring to Prof. Kreil 's discussions of his observations at Prague ( Versuch den Einfluss des Miondes auf den atmospherischen Zustand unserer Erde aus einjhhrigen Beobachtungen zu erkennen , 1841 ) .
The semiannual curves of the lunar-diurnal variation of the barometer at Prague and Melbourne closely correspond during the months from April to September , and from October to March , which seems to point to some cause common to the whole globe in a similar manner , as we know it to be with respect to the extent of the rise of the oceanic tides at the time of the solstices and equinoxes .
It would be premature to enter now upon any speculation with a view to bring the results of our observations in accordance with theory , there being still by far too few discussions on atmospheric tidal observations at our command .
The yearly curve of the lunar-diurnal variation presents some peculiarly interesting features , differing in some respects from the results of similar inquiries instituted by General Sabine and Capt. C. M. Elliot with special regard to the lunar atmospheric tides at St. Helena and Singapore , although the plan of discussion was the same .
The lunar horary variation of the barometer is as follows , " if we arrange the results in such manner that the hours are combined in which the moon is similarly situated in respect to the meridian " ( Sabine 's paper " On the Lunar Tides at St. Helena " ) : Iorary variation Mioon 's M 1oon 's distance from the obfrom the obdistance from the servations at I servations at from the meridian .
At the hours following the At the hours preceding the th ours te hours ean .
meridian .
meridian passage .
meridian passage .
following the preceding C meridian the meridian passage .
passage .
1oIh ih in .
in .
h in .
in .
in .
in .
in .
h0{ oC00ooo39 0 ?
039 +0-00080 o { I }z ?
+00002 sz 09oooz ?
+ Io -00057 oo o. .
10 ' + -000900070 I 1}0Z +i 000700140 2 -00033 Cozo09 + 'OOIZ } 2{4 00075 ) 041 zz + 000394 } ooo64 + 00014 + '00133 0+ 73 23 0055 21 600063 9 -0000 300 I5 0 00 I 'O000C7 I ooo63 -+ +§§ +o'o o69| 1}16+ -co3 + 300017 so oog 17z + 00051I + 006 oo4 6 15 , 1 } } --000 + 0009 45 5I6{6+ oo 4--6 } oooc 6+ 'oooAl c+ *0 + ooo6 + *ooo6z i6 § 4 6.00000 + 005 0 '000 ?
04.0o6z 6_ IS-04I 'o-o045 'Jr ' n both cases , at the hours following , and at those preceding the meridian passage , the minimum is decidedly at the 3ra hour , while the maximlun in the first series occurs at the Oth , and in the second at the 1st hour .
At Singapore and St. Helena , both within the tropics , the lunardiurnal variation shows a maximum at the 0th and a minimum at the 6th hour .
The discussions , based on observations made at Prague , and already referred to above , exhibit a greater conformity in respect to the lunar tides at Mielbourne than any of the tropical stations .
This conformity is especially clearly expressed in the series for " the hours following the meridian passage " ( which series seems to present in each respective case the greatest reliability ) , and we observe that the minimum occurs at the 3rd and 4th , and the maximum at the 6th hour .
But in turning Prof. Kreil 's labours in this direction to account , we must remember that they refer to a period of only one year , and cannot be considered as presenting great guarantees for decisive results , especially when considering that so high a latitude as 50§ 8 ' N. would rather have required a longer period of observation than is necessary to prove the existence and character of the lunar atmospheric tides within the tropics .
So very few discussions on this topic being at our command , it is nevertheless of considerable interest to compare the results for Prague with those at St. Helena , Singapore , and Melbourne , as done in the following little Table : Mean of three Mean of two Mean of five Mean of one years at years at years at year at Singapore St. 1ielena Melbourne Prague ( +§ 19 ' ) .
( -5§ 57 ' ) .
( -37§ 48 ) .
( +500 8 ' hI in .
in .
in .
in .
ho -0ooo570 +o-oo365 +-o'o0o o'oooooo0 0Ii 4oo0075 'oo336 'ooI2I '00043 I2 4'00330 4 '00275 -.+ '00073 -'00o oI3'+ 0080 0oo015 ooooo -0 '00039 34 +'ooI45 'oiIo +'ooo6 + 'o00005 45 +'00035 + '00046 '00045 o00032 56 '00000 oooo + -ooc62 + 000ooo7 6 Mean ... ... 4oz26z2 i -O01843 + o'oc63I + 0ooo396 Mean .
The decrease in extent of oscillation , as we recede from the equator , is clearly illustrated by the mean values of this Table .
Speaking of the extent of the oscillations , it is of importance to add a few facts relative to the amplitude , as resulting from the monthly curves .
We have seen , in the course of this exposition , that the amplitude for the semiannual periods from April to September , and from October to Marcl , is respectively 0"'00653 and 0"'00432 , which result will be materially altered in case we consider only the single months ; for inasmuch as the sense of oscillation varies considerably in the single months , constituting a semianlnual period , chiefly during summer , the combination of the hourly values of six months in one group must lecessarily tend to diminish , or even abolish in some cases , the lunar-diurnal variation .
The mean amplitude of the lunar-diurnal variation of atmospheric pressure for the several months , as represented by the means of five years , is as follows : April 0-0069 Miay 0 0171 June 0-0165 Oct. '0091 Nov. 0-0311 Dec. -0222 Means for two months equi } 0080 2l 0 distant from the equinox July 0-0131 Aug. 0-0109 Sept. 0-0133 Jan. -0108 Feb. '0093 Mar. '0139 Means for two months equi- } 0199 0101 0136 distant from the equinox The semiannual means are , for the six winter months ( when the sun 's declination is north ) , 0"`0129 , and for the summer months ( when the sun 's declination is south ) O'l-0161 , and therefore the mean amplitude in lunardiurnal variation of the barometer is 0'"0 145 .
There is evidently a great conformity in the change in extent of oscillation observable , when we examine the semiannual values of the above series .
In April and October the amplitude reaches a minimum value , whilst in the months immediately following a maximum occurs .
For both the equinoctial months the value in question is nearly alike , making at the same time the nearest approach to the annual mean .
The months following the equinoxes exhibit the smallest range in lunar-diurnal variation of atmospheric , pressure , whilst those months preceding the solstices are to be considered as maxima with respect to the value at issue .
With a view to ascertain whether the difference in the extent of the lunar atmospheric tide at the epochs of apogee and perigee may be proved to be perceptible in as high a latitude as 37§ 48 ' , I followed a course differing in some respects from the one proposed by General Sabine in his discussions of the St. Helena observations .
We have seen that in the case under consideration the hours of the extremes in pressure are not marked in a like distinct manner as for places near the equator , and I thought it on this account preferable to abandon the adherence to certain hours of the lunar day in determining the range in the value b-b , simply adopting this range for the lunar day near the apogee or perigee , irrespective of any hour of maximum or minimum .
In order to increase the number of comparisons , this range was determined in addition to the days of apogee and perigee for the day preceding and that following those epochs .
The difference in the lunar-diurnal range in atmospheric pressure at the epochs of perigee ( Rp ) and apogee ( R " ) was consequently in each case derived from six days ' observation .
Thus we obtained the following values for IRRI , which , however , cannot be immediately compared , in respect to the amount , with the corresponding values of the discussion on the St. Helena observations just referred to and arrived at by a different process .
Lunar-diurnal range in Perigee minus lunar-diurnal range in Apogee .
Months .
Mean for 1858-59 .
1859-60 .
i86o-6i .
I86i-6z .
i862-63 .
I858-63 in .
in .
in .
in .
in .
in .
April ... ... ... ... ... ... +0'775 +o'I793 +0o'910 -0 0984 +0oIoIo +0o090i8 May ... ... ... ... ... ... 'II46 '2270 '0220 '0676 '1363 '03182 June ... ... ... ... ... .
'025 '0376 + '0960 + '0380 '0837 '00248 July ... ... ... ... ... ... '79 '0786 0020 '0044 '0370 '01246 August ... ... ... ... ..'0148 + '0217 -'0750 'Io44 ?
'0587 + '1I904 September ... ... ... .'o028 '0144 ?
'1070 'oi8o '0417 +'1968 October ... ... ... ... ... oo004 + '1506 + '0400 '200o 6 1820 -z 'I5477 November ... ... ... ... '0287 '0320 oi6o + '0360 '0870 '00514 December ... ... . .
4 'I439 '0296 ?
'0760 '1ooo + '0020 + '1I846 January ... ... ... ... ... '0097 '0083 '0805 + '0177 'o27o + '01452 February ... ... ... .
'0815 '0570 'ooio '0727 + '05I '05224 March ... ... ... ... ... ... '0053 'I217 'c486 '0410 + 'oo60 + 'oo868 Means ... ... ... ... ... + '04418 + '02747 + '03605 '02808 + '00195 + '02746 Number perigee. .
13 13 14 I3 13 66 Sum .
epochs of ogee ... I4 I3 I3 13 14 67 Sum .
The mean value of + '02746 was derived with due regard to the number of epochs of apogee and perigee occurring in the whole period of observation , the total number of barometrical readings from which it was derived being 720 .
There can hardly exist a doubt , after having examined the above results , that the lunar-diurnal range in pressure of air at the time of the perigee exceeds the one at the apogee , a fact which is also in strict accordance with theory .
But it ought to be pointed out that during the months of May , June , and July the reverse seems to take place , as is manifested in every one of the five years of observation .
Whether this bears any reference to the time of aphelion on the 3rd of July , and the time of the perihelion on the 2nd of January , we do not pretend to decide now ; suffice it to have directed the attention of those more immediately interested in inquiries of this nature to a matter replete with so much interest , but as yet , comparatively speaking , scantily examined .
The mean range for the epochs of perigee and apogee is respectively 0 " 16327 and 0"'13581 , re .
sulting a general mean range of 0"'149540 .
A similar plan to that just described was pursued , in order to ascertain whether there existed any .
perceptible difference in atmospheric pressure in the periods of syzygy and quadrature .
The range of the atmospheric pressure during a lunar day was determined for days of full and change , and also for each of the epochs of quadrature separately , and furthermore for the day preceding and following each of the several epochs .
Subsequently a mean value was derived by combining the daily range of the epochs of syzygy ( Rs ) and that for the epochs of quadrature ( Rq ) .
Lunar-diurnal range in pressure of air .
[ j i Mean for the Difference , Months .
| Stl p ; ^W |il8 a epochs of Rl-Rq .
Full New First epochs OLf R-qt moon .
moon .
quarter .
quarter .
g QuadraSyzygy ' .
ture .
in .
in. .
in .
in .
i. in .
in .
April ... ... ... o'1507 0 o1632 o 1723 0-1224 0o5695 0o'4760 +0'00935 May ... ... ... '131z '1730 I '669 1399 15210 '15340 '00130 June. .
'1I25 '146 '1557 '1282 ' 12940 '14195 01255 July ... ... ... 1656 '31 1451 145 I4 4835 '12980 + '01855 August ... . .
'532 1585 1 '095 '1872 '5585 '14835 + '00750 September .
'2103 '1667 '1497 '2071 '18850 '1784 ?
0 -t '0ooo October ... . .
'1597 '2038 '797 i '557 1 '1875 '16770 j'o1405 Novemnber ... '1318 '1582 1'I648 '22o 8 1'4500 * '19280 '04780 December '1907 '2191 '1226 'i8oo 20490 1 15130 +4 '05360 January ... ... 1729 '1287 151 2059 'I5080 '17885 'o80o5 February ... '1073 '0662 'I246 '1154 '08675 1O2000 '3325 'March ... ... ... '1262 '1509 1155 -1448 '13855 |'13015 +'00840 Means ... ... . .
'1511 '15547 I'46561 *6oi6 '1 53242 '1 53360 'oooii8 The last column of this Table shows the difference R --Rq , so that plus denotes an excess of the lunar-diurnal range at the periods of syzygy , and minus an excess at the periods of quadrature .
According to the above there is a decided minimum in the lunar-diurnal range at the time of the first quarter , while the last quarter seems to be the maximum , the time of the syzygy showing intermediate values .
The general mean would indicate an excess , though very small , in favour of the epochs of the quadrature .
On examining , however , the difference for the single months , we notice that the algebraic sign denotes for seven months an excess of the epochs of syzygy , and for five only the contrary ; further , that the greatest irregularity in respect to the signs and values prevails durings the months from November to February , when hot winds are most frequent , and the sudden changes in temperature , connected with these phenomena , cause the oscillations of the barometer to be much disturbed .
The magnitude of the values during this period ought to induce us to receive them with caution , and to consider the eight remaining months separately .
The general mean difference for the eight months , from March to October , both inclusive , .
represents an excess in favour of the epochs of syzygy of 0'00676(2 inch , a value which most probably makes a near approach to truth .
If we derive mean values of the lunar-diurnal range ror the several years of observation at the respective phases of the moon , we haveLiunar-cdiurnal range in pressure of air .
Mean for the Years Full New I First Last epoch of Rs-~ moon .
moon .
quarter .
quarter .
-- ?
II --~I II SyzyCgy .
QuadraI in .
in il .
il .
in .
in .
in .
in .
1858-59 .
o-16526 0'14492 0-14627 0-17259 0'I5506 O15943 -0'00437 1859-60 .
-I3682 '14272 '1I5817 '15354 '13977 '5585 'OI608 1860-6I .
*I4017 '16672 'I2792 'I4953 'I5344 '13872 -'01472 I861-62 .
' -12969 '15270 '14664 '19473 '14119 '17068 -'02949 1862-63 .
1 '18316 '17752 '14802 '13102 '*8034 'I3952 + '04082 Means ... ... 'I5i02 'I56916 145404 160282 " 53969 'I52843 + '00oo26 The final result of this Table shows an average excess of 0"'0011 26 in favour of the epochs of syzygy , hut an analysis of this value shows that for three years the excess is in favour of the epochs of quadrature , while but two years seem to confirm what we feel inclined to regard as the rule .
So much we are able to assert , however , that the lunar-diurnal range in pressure of air at the time of the first quarter shows a minimum , and that near the last quarter and new moon a maximum in this range seems to make itself manifest .
Although the evidence adduced in the case is not of such a positive nature as that produced when treating on the question of the increased pressure of air near the perigee , we feel nevertheless inclined to believe some similar relation to exist between the atmospheric tides and the moon 's phases , as we know to be the case with respect to the oceanic tides , and that a more rigorous inquiry into this question than we are able on the present occasion to institute , will ultimately yield a result in strict accordance with the theory of gravitation .
Before concluding these researches I may be allowed to point out a fact corroborative of the result arrived at when speaking of the difference of atmospheric pressure near the epochs of syzygy and quadrature .
The mean diurnal range resulting from the last inquiry amounts to 0 " ' 153301 ; but on the former occasion we found this range to be 0"'14954 .
The excess of 0"'00376 , of which the lunar-diurnal range of the atmospheric pressure is larger , when derived from the epochs of the moon 's phases , than when obtaining it by the periods of perigee and apogee , must be attributed to the fact that in the latter case sixty-six periods of perigee were combined with sixty-seven periods of apogee , giving a fair average result ; while in the former forty-three epochs of perigee and but thirty-five of apogee happened to coincide with the several phases of the moon , tending in this way to raise the mean value of the lunar-diurnal range of the barometer above that average .

