long glass tubes, clofed at one end partly with mercury and partly with water purged of its air by boiling, and immersed the open ends in balins of mercury. On heating the upper portion of the tube, fo that the water was kept in a ftate of vapour, a bubble of air collected to the top, which continued permanent after the apparatus was removed from the fire. Plunging the tube into a long trough of mercury, he threw out this bubble, and, repeating the experiment often after intervals of days or weeks, he always obtained a new portion of air. Such is nearly the fubftance of Dr. Priestley's ftatement, which, at firft view, might appear to confirm his deductions: but, on examination, we apprehend, it will be found inconclufive. That, under the preffure of the atmosphere, water is never, by the most obstinate boiling, completely freed of the air combined with it, will hardly require demonftration. Not to wafte time in adducing arguments, we fhall only mention a fimple fact which seems to be entirely conclufive. If boiled water be expofed in a glafs to freeze, the ice will appear full of little air-bubbles, especially in the central mafs; although it be more folid and diaphanous than the ice of common water formed in fimilar circumftances. No perfon can doubt that these air-bubbles are discharged from the water previously to its congelation.We remark, by the way, that the true reafon is hence afforded why boiled water freezes moft readily, the procefs of cryftallization being in that cafe lefs retarded by the feparation of inherent air.-The air lurking in water was, therefore, fully fufficient, we prefume, to account for the refults which Dr. Priestley obtained. He has not ftated what quantity was collected of aërial fluid, nor can we infer from his expreffions that it was very confiderable. The application of heat, together with the abfence of preffure, would occafion air gradually to feparate from the water, till the quantity difengaged would, by exerting its elafticity, refift any farther extrication. This portion of air being expelled, the water, relieved from its preffure, is again brought into a condition for yielding more; and thus the fucceffive difcharges of gas will form a defcending geometrical progreffion, the terms of which are numerous or almoft unlimited.-Why did not Dr. Priestley make his experiments on a larger scale? A jar filled with hot water, and placed under the receiver of an air-pump, was likely, for inftance, to afford better profpects of fuccefs. If the Doctor's pofitions were juft, a copious production of air would enfue on a certain degree of rarefaction, and would cause the mercury to mount rapidly in the gage. It would require continued exhauftion to keep the mercury of the fame height. From our recollection of analogous obfervations, we are convinced that, with water carefully purged of its air, no fuch effects would take place:but, in reality, though Dr. Priestley's experiments were more direct and precife than we judge them to be, we should ftill hefitate to acquiefce in conclufions which are irreconcileable with every hypothefis, and repugnant to thofe general principles which refult from an extenfive train of indifputable facts.

O. P. fays that our account of the Doctor's experiments is extremely incomplete; and that it is great injustice to ftate all the author's conclufions, and not half of the facts adduced in their fupport. We are not confcious of unfairness in any part of our conduct. We are conftrained to obferve as much brevity as poffible, and our principal object was to exhibit the chain of argument. We are ever disposed to set a due value on experimental facts: but we can efteem thofe only to be of effential importance which lead to general principles. Nothing is more difgufting than a circumftantial detail of the unfuccefsful or inconclufive operations of the laboratory. It is defirable that authors fhould moderate their vanity, and fpare the patience

of their readers, by restricting their relation of experiments to the aecurate defcription of fuch as are complete and decifive. When dubious or imperfect experiments are made public, they should be ranged under the comprehenfive head of hints and conjectures.

+++ We cannot afford R. P. any farther elucidation of the calculations refpecting the population of Petersburg, which came from a correspondent on the Continent, with whom we have now no communication.

R. P. obferves that Mr. Edwards, in his Hiftory of the Weft Indies, (fee our Rev. for September, p. 68.) fpeaks of the high duties impofed on Indigo; whereas, by an act of the 7th Geo. II. c. 18. the powers of which have been conftantly renovated ever fince, the importation of that article is permitted, free of duty.-With respect to R. P.'s remark on the paffage relative to the almond and the cocoa, it appears to us that he entirely misconceives Mr. E.'s meaning, which is fufficiently obvious.

This correfpondent also remarks that Mr. Buchanan's calculations relative to the deftruction of herrings by the Solan geefe, (See Rev. for January, p. 46.) are erroneous in the refult. 100,000 Solan geefe, each devouring five herrings daily for feven months or 210 days, will in the end have eaten 105,000,000, not one hundred thousand millions.

| T. G., the tranflator of the Tour to the Pennine Alps, (fee Rev. for January, p. 62.) informs us, in juftification of his having omitted the name of the original author, that he purchased that work and the Defcription of Nice, of Mr. Beaumont, four years ago. Surely that is no reason for withholding from the public the circumstance which muft give authenticity and value to an expenfive work. We shall correct in the errata the two mistakes juftly pointed out by T. G.

tt We would gladly oblige Y. Z., but we find that applications for advice in the choice of books intrude too much on our time, and are wholly foreign from our duty.

St If S. C. be a conftant reader,' he should know that we have often stated it to be our rule not to infert criticisms on publications, which come from anonymous and unknown correfpondents. If, however, a perufal of the book in queftion juftifies S. C.'s account of it, we fhall attend to the local hints which he has given.

SS We regret that S. T.'s very polite letter came too late for acknowlegement in our laft number, as he favoured us with no other mode of addreffing him. We fhall be happy if he will enable us to give a perfonal-at leaft a private-answer.

+*+ A correfpondent, figning E. 5. controverts the account lately given of the fource of the Thames: but can we publish a contradiction of it, afferting another statement of this dubious matter to be the true one, on the authority of a letter figned E. 5.?

In the Review for January, p. 63 and 4. for Glacier de Bois, read Glacier des Bois. P. 64. 1.9 from bottom, dele (l’Arve.)

Review for February, p.171. 1. 27. read, of that fenfe. P. 225. 1. 19. dele the comma after yet.

THE

MONTHLY REVIEW,

For APRIL, 1795.

THE

ART. I. The Courfe of Hannibal over the Alps afcertained. By John Whitaker, B. D. Rector of Ruan Lanyhorne, Cornwall. 8vo. 2 Vols. pp. 670. 12s. Boards. Stockdale. 1794. "HE character of Mr. Whitaker for learning and ingenuity is well established; and the fubject of the prefent work is highly interefting, on account both of the great reputation of Hannibal, and of the fingular nature of the country through which he marched, and which exhibits, at every step, scenes of fublimity and wild magnificence: filling the mind with a kind of pleafing terror, and feeming to be a fuitable accompaniment in the train of a warrior, whofe forefight provided against every poffible contingency, whofe vigour furmounted every obftacle, and whofe ambition afpired to the conqueft of the world.

It has been obferved that the first books, which we read with attention and pleasure, influence our taste and opinions ever afterward. There is fcarcely a school-boy who has not been attracted by the atchievements of Hannibal; and the attachment which we feel for the great names of antiquity has fomething in it of the nature of friendship: we are interested in the most minute particulars relating to them; and, by an easy affociation of ideas, we are apt to regard with refpect, and with reverence, even the places which have been the scenes of their most remarkable actions.

It appears to be fomewhat fingular that, notwithstanding the notoriety of Hannibal's march over the Alps, the two great hiftorians Polybius and Livy, who record that extraordinary event, should differ very materially in the route which they affign to him. Later authors have written copiously on the subject, and, as is ufual in queftions of this fort, have formed themselves into parties; fome following Livy, and others Polybius, as their leader: but Mr. Whitaker, who feems to have entered deeper into the inquiry than any of his predeceffors, does not yield himself, implicitly to the guidance of either of thofe hiftorians. He endeavours to fupport his opinion by

VOL. XVI.

C c

matters

matters of fact, and, where those fail, by probable conjecture :: how far he has fucceeded, the reader may, in fome degree, be enabled to judge by the extracts which we fhall lay before him: Mr. W.. informs us that-

An officer of our own army, who is at once an antiquary, a foldier, and a critic, the celebrated General Robert Melvill, in 1775 took pains to trace the route of the Carthaginians, one General investigating the course of another, by an actual furvey of the ground, through the vallies and over the crests of the Alps. I am ambitious, therefore, of following the example of this amiable and friendly officer, who has moft obligingly imparted the fubftance of all his notices to me; but of following it in a different manner. I wish not to ftruggle in reality through the rugged Gullies, and to ftrain in reality up the fteep afcents with him. I mean to act on an easier, and (I think) a more effectual plan, taking the hiftories of Hannibal into my hands: comparing them with the accounts of the Roman geographers and modern travellers; collating all again with incidental notices, in other hiftorians among the ancients or among the moderns; and then delineating the courfe of the Carthaginians from the whole.

• Nor will there be found, I truft, fuch a real uncertainty in their courfe, as the difputes of the moderns and the ancients feem to announce. The generality of mankind think little on any fubject; even scholars are more apt to draw out their ftores of learning, than to exert their powers of intellect. They frequently think as little as the mereft of the mob, and my reader, who expects to walk only in the fhades of twilight, or under the glimmer of a few stars, will be agreeably furprised, I truft, to find clear light breaking in upon him, growing ftronger and ftronger as he advances, and at last forming a.. full blaze of brightness.

I first prefent myself as a guide to the Carthaginians, on the banks of the Rhone in Languedoc: here Hannibal paffed this rapid river, but at what particular point did he pass it? he had marched from the Pyrenees; not along the grand road, which we fee the Romans afterwards ufing across the fouth of France; but along another, that was higher up in the country, and came to the Rhone at a greater diftance from the fea. Almost all our knowledge of western Europe, is derived from the monuments of the Romans; and the roads of the Romans especially are our principal directors to the roads of the natives before them. That of the Romans led from the Pyrenees, to Narbonne, to Nifmes, and to Arles: this laft town was at the mouth of the Rhone, while Hannibal croffed the river almost four days march above. Hannibal, therefore, took a road to the north of this.. One accordingly occurs among the Romans, that went over the Rhone at Vienne by a bridge, of which fome appearances remain to this day. Yet this was too far to the north. Hannibal was only four days march from Arles in the fouth, as I have already noticed: but he was also four days march from Lyons in the north, as I fhall fhew hereafter. He was confequently about the middle point of the Rhone betwixt both.-Now we have one Iter of the Romans, which gives us the distance on the road between Arles and Valence, and another 8 which

which measures equally the road from Valence to Lyons. The for mer carries us from Arles to Avignon, by two intermediate ftages, twenty-three miles; to Orange, by one ftage, twenty, and to Valence, by five, feventy-one; in all one hundred and fourteen. The latter conducts us from Valence, through feventy-one miles, to Lyons: but thefe Iters obviously carry us off from the courfe of the Rhone, and lengthen the road greatly by diverting wide to the right. The real distance from Lyons to Arles, is about one hundred and fixty miles; and the middle point betwixt them, will fix us about eighty from each. This reafoning is decifively confirmed by Polybius, who ftates the place of Hannibal's paffage over the Rhone, to be seventyfive below Lyons. We muft therefore take our station many miles to the fouth of Valence; which in one of those winding Iters is feventy-one below Lyons, but in reality is about fifty-four only; and at Lauriol, near twenty miles to the fouth of Valence.'

At Lauriol in Dauphiny, then, Hannibal croffed the Rhone, and from this point we must now attend his army to the Alps: but Mr. Whitaker does not direct their march either by Mount Vifo, Mount Genévre, Mount Cenis, or by any ways adjoining to any of them: for he says that

• Hannibal ranged up along the eastern bank of the Rhone, towards Valence, Vienne, and Lyons. He thus left the long wall of the Alps at a distance on his right, while he kept the Rhone close to him on his left." He marched, (fays Livy,) up the current of the Rhone towards the midland parts of Gaul; not because this was the dire& road to the Alps, but because he thought the farther he advanced from the fea, the less likely he was to meet with the Romans, and he was inclined to avoid all encounters with them, before he had entered Italy." Hannibal, according to Polybius, placed his elephants and horfe in the rear of his army, " and advanced at the head of them along the river, marching off from the fea, and pufhing, as it were, for the midland parts of Europe." Thefe paffages are clear and peremptory, precluding all poffibility of fuppofing, if we mean to be directed by hiftory, that he left the Rhone, that he pushed directly for the borders of Gaul, and the barrier of the Alps, and that he croffed either Mount Cenis, Mount Genevre, Mount Vifo, or any other adjoining mountains, at all.'

• Hannibal now marched by Vienne to Lyons. This he reached on the fourth day from his paffage over the Rhone. He therefore marched very expeditiously, in order to leave the Romans further behind him. He actually fhews his apprehenfions of their following and overtaking him, by inverting the ufual order of his march, in ftationing those elephants and that cavalry for his rear, which at other times he ordinarily placed for his van. "He thus came to an ifland, (fays Livy,) where the Arar and the Rhone, running down from different parts of the Alps, and comprehending a portion of ground between them, unite together: to this ground they give the name of island." "He came, (adds Polybius,) to what is called an island, a region very populous and fruitful in corn, deriving its appellation from its circumstances; as here the Rhone, and there what is denominated the