fhews a gradual condenfation, or increasing brightness, towards a center or certain point; whether the outward fhape of fuch clusters or nebulæ be round, extended, or of any other given form. What has been faid with regard to the doctrine of chance, will of courfe apply to every cluster, and more efpecially to the extended and irregular fhaped ones, on account of their greater fize: It is among these that we find the largest affemblages of ftars, and most diffufive nebulofities; and therefore the odds against fuch affemblages happening without fome particular power to gather them, inareafe exceedingly with the number of the ftars that are taken together. But if the gradual accumulation ei ther of ftars or increafing brightnefs has before been admitted as a direction to the feat of power, the fame effect will equally point out the fame caufe in the cafes now under confideration. There are befides fome additional circumstances in the appear ance of extended clusters and nebulæ, that very much favour the idea of a power lodged in the brightest part. Although the form of them be not globular, it is plainly to be feen that there is a tendence towards fpherici ty, by the fwell of the dimenfions the nearer we draw towards the moft luminous place, denoting as it were a courfe, or tide of ftars, fetting towards a center. And if allegorical expreffions may be allowed-it fhould feem as if the ftars thus flocking towards the feat of power were ftemmed by the crowd of thofe already affembled, and that while fome of them are fuccefsful in forcing their predeceffors fideways out of their places, others are themfelves obliged to take up with lateral fituations, while all of them fecm equally to ftrive for a place in the central fwelling, and generating fpherical figure.

Since then almost all the nebula and clusters of stars I have feen, the umber of which is not lefs than three

and twenty hundred, are more condenfed and brighter in the middle; and fince, from every form, it is now equally apparent that the central accumulation or brightnefs must be the refult of central powers, we may venture to affirm that this theory is no longer an unfounded hypothesis, but is fully established on grounds which cannot be overturned.

Let us endeavour to make fome ufe of this important view of the conftructing canfe, which can thus model fidereal fyftems. Perhaps, by placing before us the very extenfive and varied collection of clusters, and nebulæ furnished by my catalogues, we may be able to trace the progrefs of its operation, in the great laboratory of the Universe.

If thefe clusters and nebula were all of the fame fhape, and had the fame gradual condensation, we should make but little progrefs in this inquiry; but, as we find fo great a variety in their appearances, we shall be much fooner at a lofs how to account for fuch various phænomena, than be in want of materials upon which to exexcife our inquifitive endeavours.

Some of these round clusters confit of ftars of a certain magnitude, and given degree of compreffion, while the whole clufter itself takes up a space of perhaps 10 minutes; others appear to be made up of stars that are much fmaller, and much more compressed, when at the fame time the cluster itfelf fubtends a much smaller angle, fuch as 5 minutes. This diminution of the apparent fize, and compreffion of itars, as well as diameter of the clufter to 4, 3, 2 minutes, may very confiftently bo afcribed to the different diftances of these clusters from the place in which we obferve them; in all which cafes we may admit a general equality of the fizes and compreffion of the ftars that compofe them, to take place. It is alfo highly probable that a continuation of fuch decreafing magnitudes, and increafing

com

compreffion, will justly account for the appearance of round, eafily refolvable, hebulæ; where there is almost a certainty of their being clusters of ftars. And no Aftronomer can hesitate to go ftill farther, and extend his furmifes by imperceptible fteps to other nebula, that ftill preferve the fame characteriftics, with the only variations of vanishing brightness, and reduction of fize.

fpherical clufters. Any force, that acts uninterruptedly, muft produce effects proportional to the time of its action. Now, as it has been fhewn that the spherical figure of a cluster of ftars is owing to central powers, it follows that thofe clufters which ce teris paribus, are the most compleat in this figure, muft have been the longeft exposed to the action of these caufes. This will admit of various points Other clusters there are that, when of views. Suppofe, for instance, that they come to be compared with fome 5000 ftars had been once in a certain of the former, feem to contain ftars of fcattered fituation, and that other 5000 an equal magnitude, while their com- equal ftars had been in the fame fituapreffion appears to be confiderably tion, then that of the two clusters different. Here the fuppofition of which had been longest exposed to the their being at different distances will action of the modelling power, we either not explain the apparently fuppofe, would be most condensed, greater compreffion, or, if admitted to and more advanced to the maturity of do this, will convey to us a very in its figure. An obvious confequence ftructive confequence: which is, that that may be drawn from this confiderthe ftars which are thus fuppofed ation is, that we are enabled to judge not to be more compreffed than thofe of the relative age, maturity, or cliin the former clufter, but only to ap- max of a fidereal fyftem, from the pear fo on account of their greater difpofition of its component parts diftance, muft needs be proportion and, making the degrees of brightally larger, fince they do not appear of nefs in nebulæ ftand for the different lefs magnitude than the former. As accumulation of ftars in clusters, the fame conclufions will extend equally to them all. But we are not to conclude from what has been faid that every fpherical clufter is of an equal ftanding in regard to abfolute duration, fince one that is compofed of a thousand ftars only must certainly arrive to the perfection of its form fooner than another, which takes in a range of a million. Youth and age are comparative expreffions; and an oak of a certain age may be called very young, while a cotemporary fhrub is already on the verge of its decay. So that, for inftance, a cluster or nebula which is very gradually more compreffed and bright towards the middle, may be in the perfection of its growth, when another which approaches to the condition pointed out by a more equal compreffion, fuch as the nebula I have called Planetary feem to prefent us with, may be look

therefore, one or other of thefe hypothefes must be true, it is not at all improbable but that, in fome inftances, the ftars may be more compreffed and in others, of a greater magnitude. This variety of fize in different fpherical clusters, I am however inclined to believe, may not go farther than the difference in fize, found among the individuals belonging to the fame fpecies of plants, or animals, in their different ftates of age, or vegetation, after they are come to a certain de gree of growth. A farther inquiry into the circumftance of the extent, both of condenfation and variety of fize, that may take place with the ftars of different clufters, we fhalf poftpone till other things have been previously difcuffed.

Let us then continue to turn our view to the power which is moulding the different affortments of tars into VOL. XI. No. 61.

D

ed

ed upon as very aged, and drawing on towards a period of change, or diffolution. This has been before furmifed, when, in a former paper, I confidered the uncommon.degree of compreffion that must prevail in a nebula to give it a planetary afpect; but the argument, which is now drawn from the powers that have collected the formerly fcattered ftars to the form we find they have affumed, muft greatly cor roborate that fentiment.

and one advantage we may at leas reap from it is, that we can, as it were, extend the range of our experience to an immenfe duration. For, to con tinue the fimile I have borrowed from the vegetable kingdom, is it not almoft the fame thing, whether we live fucceffively to witnefs the germination, blooming, foliage, fecundity, fading, withering, and corruption of a plant, or whether a vast number of fpecimens, felected from every ftage through which the plant paffes in the course of its existence, be brought at once to our view?.

This method of viewing the hea vens feems to throw them into a new kind of light. They now are feen to refemble a luxuriant garden, which contains the greatest variety of productions, in different flourishing beds; Slough near Windfor, May 1, 1789.

WILLIAM HERSCHEL.

On the Harmony of the Dead Languages. From Alembert's Thoughts, Maxims, and Principles.

WE every day hear men of letters

expatiating on the harmony of the Greek and Latin languages, and on the fuperiority of thefe over the modern languages, without reckoning other advantages ftill greater, which they attribute to the nature and genius of the former. Admiration for the harmony of the dead and learned languages, is particularly remarkable in those who, having spent much time in ftudying them, are apt to flatter themselves, that they understand them well, and who, perhaps, do understand them as well as it is poffible for a dead language to be understood, that is, very imperfectly.

This enthufiafm, which is not always as genuine as it appears, originates in a very pardonable felf-love. After having been long employed in the ftudy of a language, we are unwilling to think that our time has been thrown away, we even wish to feem, in the eyes of others, amply rewarded for the pains we have taken, and we often fay with affected tranf fport: Ab if you but underfood Greek!

Thofe who understand, or think they understand the Hebrew, the Arabic, the Syriac, the Coptic, the Perfian, the Chinefe, &c. think and fpeak in the fame way, and for the fame reafon. The language they have learnt is always the most beautiful, the most copious, the most harmonious ; juft as a dependant always thinks his patron a fuperior man. But when the merit of a man in power is ex pofed to public view, the praises be towed on him, if he is unworthy of them, are fcornfully difallowed by the public; while the languages that are called learned, being almoft entirely unknown, expofe their panegyrifts to no danger of contradiction; as those who are in a capacity of doing fo are. equally interested in preaching up the value of the object of their study and attention.

But the modern Greek and Latin scholars are not quite at their ease; as many befides themselves have a finattering of Greek, and a reasonable acquaintance with the Latin tongue, it is eafy to embarrals them on the

subject

fubject of their admiration.

It is faid, for example, the French, the English, the Germans, and the Italians, pronounce the Latin language very differently from each other, in fo much, that they scarce understand one another when pronouncing it, and you would hardly ima gine they were speaking the fame language: all of them think it very harmonious; but can they all be serious fince, properly speaking, it is not the fame language they are pronouncing? and does it not therefore follow, that that pretended harmony, which the modern admirers of the Latin tongue, exait fo much, exifts at least as much in their imagination as in their ears? In order to decide this question, as well at least as it is in our power to do, we must firit of all examine what is understood, or ought to be understood, by the harmony of a language; ave must then determine in what the harmony of a dead language can confit, with regard to us, and efpecially what that harmony is in the Latin Janguage, which of all the dead ones, is the most familiar and best known.

Let us, first of all, take notice, that what is called the harmony of a language, ought rather to be called its melody. For harmony is properly the pleasure arifing from feveral founds heard at once; melody is that which refults from a fucceffion of fingle founds: now, what is called the harmony of a language, is the pleasure which refults from the fucceffion of founds in a difcourfe pronounced in that language; and it would, therefore, be better to give this pleasure the name of melody: but this by the bye; we will abide by the terms in -common ufe, after having fettled the -precife idea which belongs to them.

then with fyllables, which are formed either of fimple vowels, or of confonants joined with vowels. Now, of vowels and confonants fome are more eafily pronounced than others, they are more or less mute, more or lefs harfh, and it is the combination of thefe vowels and confonants which makes a fyllable more or lefs foft, more or less dull. Befides, as there are fyllables which are more or lefs eafily pronounced, there are also fucceffions of fyllables which are more easily pronounced than others. One fyllable is more cafily or with more difficulty pronounced after another, according as the organs of pronunciation preferve the difpofition they were in at pronouncing the firit, or are obliged to change it. On which account we must remark, that two confecutive confonants form each a syilable, because there is always neceffarily a mute é interpofed between them; and as this e mute paffes quickly, and is not pronounced, the organs are obliged to make the greater effort to mark the double confonant. This is the reafon why languages, fuch as the German, which abound in confonants following close upon one another; are more harsh than those languages in which the confonants are not fo frequent.

A language abounding in vowels, and efpecially in foft vowels, like the Italian, would be the fofteft of all; it would not be the most harmonious, for melody, in order to be agreeable, must not only be foft, but varied. A language which should have, like the Spanish, a happy mixture of vowels and of foft and fonorous confonants, would perhaps be the most harmonious of the living and modern languages. Melody in ipeech has a good deal of analogy with mufical melody. An air in which only the intervals of the diatonie fcale fhould be employed, would be languid; an air that should contit only of the most perfect concords, fuch as the third and fifth, would be D 2

In order to analyfe the pleasure which refults from a fucceffion of founds, we must decompofe that fucceffion into its parts and elements: thus, phrases are compofed of words, and words of fyllables. Let us begin

infipid