An Easy Demonstration of the Analogy of the Logarithmick Tangents to the Meridian Line or sum of the Secants : with various Methods for computing the same to the utmost Exact ness , by E. Halley .
IT is now near 100 Years since our Worthy Countryman Mr. Edward Wright published his Correction of Errors in Navigation , a Book well deserving the perusal of all such as design to use the Sea .
Therein he considers the Course of a Ship on the Globe , stearing obliquely to the Meridian ; and having shown , that the Departure from the Meridian , is in all cases less than the Difference of Longitude , in the ratio of Radius to the secant of the Latitude , he concludes , That the sum of the Secants of each point in the Quadrant being added successively , would exhibit a Line divided into Spaces , such as the intervals of the parallels of Latitude ought to be in a true Sea Chart , whereon the Meridians are made parallel Lines , and the Rhombs or Oblique Courses represented by right Lines .
This is commonly known by the name of the Meridian Line , which though it generally be called Mercators , was yet undoubtedly Mr. Wrights Invention , ( as he has made it appear in his Preface .
) And the Table thereof is to be met with in most Books treating of Navigation , computed with sufficient exactness for the purpose .
It was first discovered by chance , and as far as I can learn , first published by Mr. Henry Bond , as an addition to Norwoods Epitome of Navigation , about 50 Years since , that the Meridian Line was Analogous to a Scale of Logarithmick Tangents of half the Complements of the Latitudes .
The difficulty to prove the truth of this proposition , seemed such to Mr. Mercator , the Author of Logarithmotechnia , that he proposed to wager a good sum of Money , against whoso would fairly undertake it , that he should not demonstrate either , that it was true or false : And about that time Mr. John Collins , holding a Correspondence with all the Eminent Mathematicians of the Age , did excite them to this Enquiry .
The first that demonstrated the said Analogy , was the excellent Mr. James Gregory in his Exercitationes Geometricae , published Anno 1668 .
which he did , not without a long train of Consequences and Complication of Proportions , whereby the evidence of the Demonstration is in a great measure lost , and the Reader wearied before he attain it .
Nor with less work and apparatus hath the celebrated Dr. Barrow in his Geometrical Lectures ( Lect. XI .
App. I. ) proved , that the Sum of all the Secants of any arch is analogous to the Logarithm of the ratio of Radius + Sine to Rad. Sine , on which is all one , that the Meridional parts answering to any degree of Latitude , are as the Logarithms of the rationes of the Versed Sines of the distances from both the Poles .
Since which the incomparable Dr. Wallis ( on occasion of a paralogism committed by one Mr. Norris in this matter ) has more fully and clearly handled this Argument , as may be seen in Num. 176 .
of these Transactions .
But neither Dr. Wallis nor Dr. Barrow in their said Treatises have any where touched upon the aforesaid relation of the Meridian-line to the Logarithmick Tangent ; nor hath any one , that I know of , yet discovered the Rule for computing independently the interval of the Meridional parts answering to any two given Latitudes .
Wherefore having attained , as I conceive , a very facile and natural demonstration of the said Analogy , and having found out the Rule for exhibiting the difference of Meridional parts , between any two parallels of Latitude , without finding both the Numbers whereof they are the difference : I hope I may be entitled to a share in the emprovements of this useful part of Geometry .
Desiring no other favour of some Mathematical Pretenders , than that they think fit to be so just , as neither to attribute my desire to please the Honourable the Royal Society in these Exercises , to any kind of Vanity or Love of Applause in me , ( who too well know how very few these things oblige , and how small reward they procure ) nor yet to complain coram non judice , that I arrogate to my self the Inventions of others , and upon that pretext to depretiate what I do , unless at the same time , they can produce the Author I wrong , to prove their assertions .
Such disingenuity as I have always most carefully avoided , so I wish I had not too much experience of it in the very same persons , who make it their business to detract from that little share of Reputation I have in these things .
But to return to the matter in hand , Let us demonstrate the following Proposition .
The Meridian Line is a Scale of Logarithmick Tangents of the half Complements of the Latitudes .
For this Demonstration , it is requisite to premise these four Lemmata .
Lemma .
I. In the Stereographick Projection of the Sphere upon the plain of the Equinoctial , the distances from the Center , which in this case is the Pole , are laid down by the Tangents of half those distances , that is , of half the Complements of the Latitudes .
This is evident from Eucl .
3.20 .
Lem .
II .
In the Stereographick Projection , the Angles , under which the Circles intersect each other , are in all cases equal to the Spherical Angles they represent : Which is perhaps as valuable a property of this Projection , as that of all the Circles of the Sphere thereon appearing Circles : But this not being vulgarly known , must not be assumed without a Demonstration .
Let EBPL be any great circle of the Sphere , E the Eye placed in its Circumference , C its Center , P any point thereof , and let FCO be supposed a plain erected at right Angles to the Circle EBP L , on which FCO we design the Sphere to be projected .
Draw EP crossing the Plain FCO in p , and p shall be the point P projected .
To the point P draw the Tangent AP G , and on any point thereof , as A , erect a perpendicular AD , at right angles to the plain EBPL , and draw the lines PD , AC , DC : and the angle APD shall be equal to the Spherical Angle contained between the plains APC , DPC .
Draw also AE , DE , intersecting D AK B LP dG FCO ap secting the plain FCO in the points a and d ; and join a d , p d : I say the Triangle adp is simular to the triangle ADP , and the angle apd equal to the angle APD .
Draw PL , AK parallel to FO , and by reason of the parallels , ap will be to ad as AK to AD : But ( by Eucl .
3 .
32 .
) in the triangle AKP , the angle AKP=LPE is also equal to APK=EPG wherefore the sides AK , AP are equal , and itwill be , as ap to ad so AP to AD .
Whence the angles DAP , dap being right , the angle APD will be equal to the angle apd that is , the Spherical Angle is equal to that on the Projection , and that in all Cases .
Which was to be proved .
This Lemma I lately received from Mr. Ab .
de Moivre , though I since understand from Dr. Hook that he long ago produced the same thing before the Society .
However the demonstration and the rest of the discourse is my own .
Lemma III .
On the Globe , the Rhumb Lines make equal angles with every Meridian , and by the aforegoing Lemma , they must likewise make equal angles with the Meridians in the Stereographick Projection on the plain of the Equator : They are therefore , in that Projection , Proportional Spirals about the Pole Point .
Lemma IV .
In the Proportional Spiral it is a known property that the angles BPC or the arches BD , are Exponents of the rationes of BP to PC : for if the arch BD be divided into innumerable equal parts , right lines drawn from them to the Center P , shall divide the Curve BccC into an infinity of proportionals ; and all the lines Pc shall be an infinity of proportionals between PB and PC whose number is equal to all the points d , d , in the arch BD : Whence , and by what I have delivered in Num. 216 , it follows , that as BD to Bd , or as the angle BPC to the angle BPc , so is the Logarithm of the ratio of PB to PC , to the Logarithm of the ratio of PB to Pc .
P C D c c Bdd From these Lemmata our Proposition is very clearly demonstrated : For by the first , PB , Pc , PC are the Tangents of half the Complements of the Latitudes in the Stereographick Projection : And by the last of them , the differences of Longitude , or angles at the Pole between them , are Logarithms of the rationes of those Tangents one to the other .
But the Nautical Meridian Line is no other than a Table of the Longitudes , answering to each minute of Latitude , on the Rhumbline making an angle of 45 degrees with the Meridian .
Wherefore the Meridian Line is no other than a Scale of Logarithmick Tangents of the half Complements of the Latitudes .
Quod erat demonstrandum .
Coroll. 1 .
Because that in every point of any Rhumb Line , the difference of Latitude is to the Departure , as the Radius to the Tangent of the angle that Rhumb makes with the Meridian ; and those equal Departures are every where to the differences of Longitude , as the Radius to the Secant of the Latitude ; it follows that the differences of Longitude are , on any Rhumb , Logarithms of the same Tangents , but of a differing Species ; being proportioned to one another as are the Tangents of the angles made with the Meridian .
Coroll. 2 .
Hence any Scale of Logarithm Tangents , ( as those of the Vulgar Tables made after Brigg 's form ; or those made to Napiers , or any other form whatsoever ) is a Table of the differences of Longitude , to the several Latitudes , upon some determinate Rhumb or other : And therefore , as the Tangent of the angle of such Rhumb , to the Tangent of any other Rhumb : So the difference of the Logarithms of any two Tangents , to the difference of Longitude , on the proposed Rhumb , intercepted between the two Latitudes , of whose half Complements you took the Logarithm Tangents .
And since we have a very complete Table of Logarithm Tangents of Brigg 's form , published by Vlacq , Anno 1633 , in his Canon Magnus Triangulorum Legarithmicus , computed to ten Decimal places of the Logarithm , and to every ten Seconds of the Quadrant , ( which seems to be more than sofficient for the nicest Calculator ) I thought fit to enquire the Oblique angle , with which that Rhumb Line crosses the Meridian , whereon the said Canon of Vlacq precisely answers to the differences of Longitude , putting Unity for one minute thereof , as in the Common Meridian Line .
Now the momentary augment or fluxion of the Tangent Line at 45 degrees , is exactly double to the fluxion of the arch of the Circle , ( as may easily be proved ) and the Tangent of 45 , being equal to Radius , the fluxion also of the Logarithm Tangent will be double to that of the arch , if the Logarithm be of Napeirs form : But for Brigg 's form it will be as the same doubled arch multiplied into , 0 , 43429 , &amp;c. or divided by 2 , 30258 , &amp;c. Yet this must be understood only of the addition of an indivisible arch , for it ceases to be true if the arch have any determinate magnitude .
Hence it appears , that if one minute be supposed Unity , the length of the arch of one minute being ,000290888208665721596154 &amp;c. in parts of the Radius , the proportion will be as Unity to 2,908882 &amp;c. so Radius to the Tangent of 71~ 1~ 42~ whose Logarithm is 10 .
46372611720718325204 &amp;c. and under that angle is the Meridian intersected by that Rumb Line , on which the differences of Napeirs Logarithm Tangents of the half Complements of the Latitudes are the true differences of Longitude , estimated in minutes and parts , taking the first Four Figures for Integers .
But for Vlacq 's Tables we must say .
As.2302585 &amp;c. to 2908882 &amp;c. So Radius to 1,26331143874244569212 , &amp;c. which is the Tangent of 51~ 38~ 9~ , and its Logarithm 10,101510428507720941162 &amp;c. wherefore in the Rhumb Line , which makes an angle of 51~ 38~ 9~ with the Meridian , Vlacq 's Logarithm Tangents are the true differences of Longitude .
And this compared with our second Corollary may suffice for the use of the Tables already computed .
But if a Table of Logarithm Tangents be made by extraction of the root of the Infiniteth power , whose Index is the length of the arch you put for Unity , ( as for minutes the,0002908882th &amp;c. power ) which we will call a ; such a Scale of Tangents , shall be the true Meridian Line or sum of all the Secants taken infinitely many .
Here the Reader is desired to have recourse to my little Treatise of Logarithms , published in N~ 216 .
p. 58 .
that I may not need to repeat it .
By what is there delivered , it will follow , that putting t for the excess or defect of any Tangent above or under the Radius or Tangent of 45 ; the Logarithm of the ratio of Radius to such Tangent will be 1/ m into t 1/ 2tt + 1/ 3ttt 1/ 4tttt + 1/ 5t~ , &amp;c. when the arch is greater than 45gr , or 1/ m into t+ 1/ 2tt + 1/ 3t3 + ~t4 + 1/ 5t5 , &amp;c. when it is less than 45g~ .
And by the same doctrine put ting T for the Tangent of any arch , and t for the difference thereof from the Tangent of another arch , the Logarithm of their ratio will be 1/ m into t/ T + tt/ 2TT + t3/ 3T3 + t4/ 4T4 + t5/ 5T5 , &amp;c. when T is the greater Term , or 1/ m into t/ T tt/ 2TT + t3/ 3T3 t4/ 4T4 + t5/ 5T5 , &amp;c. when T is the lesser Term : And if m be supposed,0002908882 , &amp;c. = a , its reciprocal r/ a will be , 3437,7467707849392526 , &amp;c. which multiplied into the aforesaid Series , shall give precisely the difference of Meridional parts , between the two Latitudes to whose half complements the assumed Tangents belong .
Nor is it material from whether Pole you estimate the Complements , whether the elevated or depressed ; the Tangents being to one another in the same ratio as their Complements , but inverted .
In the same Discourse I also showed that the Series might be made to converge twice as swift , all the even Powers being omitted : and that putting ~ for the sum of the two Tangents the same Logarithm would be 2/ m or 2r/ a into t/ ~ + t3/ 3~3 + t5/ 5~5 + t7/ 7~7 + t9/ 9~9 , &amp;c. but the ratio of ~ to t , or of the sum of two Tangents to their difference , is the same as that of the sine of the sum of the arches , to the sine of their difference .
Wherefore if S be put for the sine Complement of the Middle Latitude , and s for the sine of half the difference of Latitudes , the same Series will be 2 r/ a into s/ S +s 3/ 3S3 +s 5/ 5S5 +s 7/ 7S7 +s 9/ 9S9 , &amp;c. wherein as the differences of Latitude are smaller , fewer steps will suffice .
And if the Equator be put for the Middle Latitude , and consequently S= R , and s to the sine of the Latitude , the Meridional parts reckoned from the Equator will be s/ a +s 3/ 3rra +s 5/ 5r4a +s 7/ 7r6a , &amp;c. which is coincident with Dr. Wallis 's solution in Numb. 176 .
And this same Series , being half the Logarithm of the ratio of R+s to R s , that is , of the Versed-sines of the distances from both Poles , does agree with what Dr. Barrow had shown in his XI .
Lecture .
The same ratio of T to t may .
be expressed also by that of the Sum of the Co-sines of the two Latitudes , to the sine of their difference : As likewise by that of the Sine of the Sum of the two Latitudes , to the difference of their Co sines : Or by that of the Versed-sine of the Sum of the Co-latitudes , to the difference of the sines of the Latitudes ; Or as the same difference of the sines of the Latitudes , to the Versed-sine of the difference of the Latitudes ; all which are in the same ratio of the Co-sine of the Middle Latitude , to the Sine of half the difference of the Latitudes .
As it were easy to demonstrate , if the Reader were not supposed capable to do it himself , upon a bare inspection of a Sheme duly representing these Lines .
This variety of Expression of the same ratio I thought not fit to be omitted , because by help of the rationality of the Sine of 30gr , in all cases where the Sum or difference of the Latitudes is 30gr , 60gr , 90gr , 120gr or 150 degrees , some one of them will exhibit a simple series , wherein great part of the Labour will be saved : And besides I am willing to give the Reader his choice which of these equipollent methods to make use of ; but for his exercise shall leave the prosecution of them , and the compendia arising therefrom , to his own industry .
Contenting my self to consider only the former , which for all uses seems the most convenient , whether we design to make the whole Meridian Line , or any part thereof , viz. 2 r/ a into s/ S +s 3/ 3S3 +s 5/ 5S5 +s 7/ 7S7 +s 9/ 9S9 , &amp;c. Wherein a is the length of any Arch which you design shall be the Integer or Unity in your Meridional Parts ; ( whether it be a Minute , League or Degree , or any other , ) S the Cosine of the Middle Latitude , and s the Sine of half the difference of Latitudes ; But the Secants being the Reciprocals of the Co sines , s/ S will be equal to ss/ rr putting s for the Secant of the Middle Latitude ; and 2r/ a into s/ S will be = 2ss/ are This multiplied by s s/ 2SS that is by ssss/ 3rrrr , will give the second step : and that again by 3 ssss/ 5rrrr , the third step ; and so forward till you have completed as many Places as you desire .
But the squares of the Sines being in the same ratio with the Versed sines of the double Arches , we may instead of s s/ 3SS assume for our Multiplicator v/ 3 V ' or the Versedsine of the difference of the Latitudes divided by thrice the Versed-sine of the sum of the Co-latitudes , &amp;c. which is the utmost Compendium I can think of for this purpose , and the same series will become .
2s r/ a S into 1+ v/ 3 V+ v2/ 5 V2 + v3/ 7 V3 + v4/ 9 V4 Hereby we are enabled to estimate the default of the method of making the Meridian line by the continual addition of the Secants of aequidifferent Archs , which as the differences of those Arches are smaller , does still nearer and nearer approach the Truth .
If we assume , as Mr. Wright did , the Arch of one Minute to be Unity , and one Minute to be the common difference of a rank of Arches : It will be in all cases , As the Arch of one Minute , to its Chord : : So the Secant of the Middle Latitude , to the first step of our series .
This by reason of the near equality between a and 2 s , which are to one another in the ratio of Unity to 1-0,00000000352566457713 , &amp;c. will not differ from the Secant s but in the ninth Figure ; being less than it in that proportion .
The next step being +2s3s 3/ 3 a r~ will be equal to the Cube of the Secant of the middle Latitude multiplied into 2ss s/ 3arr= 0,00000000705132908715 ; which therefore unless the Secant exceed ten times Radius , can never amount to 1 in the fifth place .
These two steps suffice to make the Meridian Line or Logarithm Tangent to far more places than any Tables of Natural Secants yet extant , are computed to ; but if the third step be required it will be found to be + ss into 2s ~/ 5 ar4 = 0,000000000000000089498 ; By all which it appears that Mr. Wright 's Table does no where exceed the true Meridian Parts by fully half a Minute ; which small difference arises by his having added continually the Secants of 1~,2~,3~ , &amp;c. instead of 01/ 2~,11/ 2,2~1/ 2 , 3~1/ 2 , &amp;c. But as it is , it is abundantly sufficient for Nautical Uses .
That in Sr. Jonas Moor 's New Systeme of the Mathematicks is much nearer the Truth , but the difference from ~ is scarce sensible , till you exceed those Latitudes where Navigation ceases to be practicable , the one exceeding the Truth by about half a Minute , the other being a very small matter deficient therefrom .
For an example easy to be imitated by whofo pleases , I have added the true Meridional Parts to the first and last Minutes of the Quadrant ; not so much that there is any occasion for such accuracy , as to show that I have obtained , and laid down herein , the full Doctrine of these spiral Rhumbs which are of so great concern in the Art of Navigation .
The first Minute is , 1,00000001410265862178 The Second , 2,00000005641063806707 The Last , or 89~~59~ is 30374,9634311414228643 and not 32348,5279 as Mr. Wright has it , by the addition of the Secants of every whole Minute : Nor 30249,8 as Mr. Oughtred 's Rule makes it , by adding the Secants of every other half Minute .
Nor 30364,3 as Sir Jonas Moor had concluded it by I know not what method , tho ' in the rest of his Table he follow Oughtred .
And And this may suffice to show how to derive the true Meridian Line from the Sines , Tangents or Secants supposed ready made ; but we are not destitute of a Method for deducing the same independently , from the Arch it self .
If the Latitude from the Equator be estimated by the length of its Arch A , Radius being Unity , and the Arch put for an Integer be a , as before ; the Meridional Parts answering to that Latitude will be 1/ ainto A+1/ 6 A3+1/ 24 A3+1/ 84/ 60 A~ or 61/ 5040 A7+-11/ 2880/ 1386 A9 or 1385/ 362880 A9 , &amp;c. which converges much swifter than any of the former Series , and besides has the advantage of Aencreasing in Arithmetical progression , which would be of great ease , if any should undertake de novo to make the Logarithm Tangents , or the Meridian Line to many more places than now we have them .
The Logarithm Tangent to the Arch of 45 + 1/ 2 A being no other than the aforesaid Series A+ 1/ 6 A3 + 1/ 24 A~ , &amp;c. in Napeirs form , or the same multiplied into 0,43429 , &amp;c. for Briggs 's .
But because all these Series towards the latter end of the Quadrant do converge exceeding slowly , so as to render this Method almost useless , or at least very tedious .
It will be convenient to apply some other Arts , by assuming the Secants of some intermediate Latitudes ; and you may for s or the Sine of ~ the Arch of half the difference of Latitudes , substitute ~ 1/ 6 ~3 + 1/ 120 ~5 1/ 5040 ~7 + 1/ 362880 ~9 , &amp;c. according to Mr. Newton 's Rule for giving the Sine from the Arch ; And if ~ be no more than a degree , a very few steps will suffice for all the accuracy that can be desired .
And if ~ be commensurable to a , that is , if it be a certain number of those Arches which you make your Integer , then will ~/ a be that number : which if we call n. the parts of the Meridional Line will be found to be .
1+ ss~~/ 3r4 + s4~4/ 5r8 + s6~6/ 7r12 , &amp;c. ~~/ 6rr ss ~4/ 6r6 s4 ~6/ 6 r20 , &amp;c. sn/ r into + 1/ 120 ~4/ r4 + 13 s2 ~6/ 360 r8 , &amp;c. 1 ~6/ 5040 r6 , &amp;c. In this the first two steps are generally sufficient for Nautical uses , especially when neither of the Latitudes exceed 60 degrees , and the difference of Latitudes doth not pass 30 degrees .
But I am sensible I have already said too much for the Learned , tho ' too little for the Learner ; to such I can recommend no better Treatise , than that of Dr. Wallis in Numb. 176 .
wherein he has with his usual brevity , and that perspecuity peculiar to himself , handled this Subject from the first Principles , which here for the most part we suppose known .
I need not show how , by regressive work , to find the Latitudes from the Meridional Parts , the Method being sufficiently obvious .
I shall only conclude with the proposal of a Problem which remains to make this Doctrine complete , and that is this .
A ship sails from a given Latitude , and having run a certain number of Leagues , has altered her Longitude by a given angle .
It is required to find the Course she steered .
The solution hereof would be very acceptable , if not to the public , at least to the Author of this Tract , being likely to open some further Light into the Mysteries of Geometry .
To Conclude , I shall only add , that Unity being Radius , the Cosine of the Arch A , according to the same Rules of Mr. Newton , will be 1 1/ 2 A2 + 1/ 24 A4 1/ 720 A~ + 1/ 40320 A8 1/ 3628800 A10 &amp;c. from which and the former Series exhibiting the Sine by the Arch , by division it is easy to conclude , that the Natural Tangent to the Arch A is A+ 1/ 3 A3 + 2/ 15 A5 + 17/ 315 A7 + 62/ 2835 A9 &amp;c. and the Natural Secant to the same Arch 1+ 1/ 2 A2 + 5/ 24 A4 + 1/ 726 A6 + 277/ 8064 A8 &amp;c. and from the Arithmetic of Infinites , the Number of these Secants being the Arch A , it follows , that the sum Total of all the Infinite Secants on that Arch is A+ 1/ 6 A3 + 1/ 24 A5 + 68/ 5040 A7 + 277/ 72576 A9 &amp;c. the which , by what foregoes , is the Logarithm Tangent , of Napeirs form , for the Arch of 45gr + 1/ 2 A , as before .
And Collecting the Infinite Sum of all the Natural Tangents on the said Arch A , there will-arise 1/ 2 AA+ 1/ 12 A4 + 1/ 45 A6 + 17/ 2520 A8 + 31/ 24175 A10 &amp;c. twhich will be found to be the Logarithm of the Secant of

