S0 much has already been written about Paris-Madrid that it would

• seem to be a work of supererogation to write any more.; but a correspondent has recently asked me a question with regard to this race, which is still so famous although it was run as long ago as May 24th, 1903, and the answer to this question involves. points which proved to be of so much interest to me that it struck me that they might be equally intriguing to others also. If in the coarse of this inquiry it becomes necessary in part to destroy a legend, I am sorry for it, and can only plead that the truth is sufficiently nutrvellous as to need no superstructure of myth in order to inspire posterity with awe.

As is generally known, therefore, Paris-Madrid, which was the last of the great town-to-town races, although always known as such, is really misleadingly socaned, because, as a result of the number of accidents on the first stage to Bordeaux, the race was stopped there by the authorities, and not one of the racing ears got nearer than that to Madrid. The course of the race, therefore, was actually Paris-Bordeaux, and the distance, after deducting the portions of the road passing through towns which were neutralised, was 342 miles. Over this distance the winner proved to be Gabriel on a 70-h.p. Mors, whose time was 5 hours 14 minutes 31.2 secoads, and whose speed was 65.8 m.p.h.

This perfOrtnance is sufficiently remarkable in itself ; but the greatest diffitulty with which the drivers in the race had to contend was the dust, with the result Mit those who started at the head of the procession gained an enormous initial advantage. The number of Gabriel's Mors, however, was 168, and quite a few cars, therefore, started in front of it. They started, in fact, at intervals of one minute, and yet—and this, quite rightly, has always aroused the wonder of posterity—Gabriel's Mom was the third car to reach Bordeaux.

The fact has, indeed, aroused so much wonder in my correspondent that he is almost inclined to doubt its possibility. For example, he argues, car No. 8 apparently started 160 minutes or 2 kthirs 40 minutes before Gabriel, and unless it took nearly 8 hours over the journey to Bordeaux, Which would be an average Of about 43 m.p.h., it must. have got there first. In fact we know that over th irty cars made a better average than 43 mph. for the journey. How was it, therefore, that only two of them reached their journey's end before Gabriel's Mors ?

If I could journey to Paris and consult the archives of the Automobile Club de Prance, I could, no doubt, ansNver my correspondent's question very accurately ; but not being in a position to do so, I must fall back on a very good second best in the shape of Gerald Rose's "Record of Motor Racing:" Without Rose's work in front of one, it may, I fear, be rather difficult to follow the present thesis. I take it, however, that anyone who is Interested in Paris-Madrid probably possesses this classic work, and anyone who is not has probably already abandoned this article. I proceed, therefore, with a comparatively clear conscience. Now the first point is that my correspondent hits been misled by the fact that

Gabriel's Mors carried the number 168 into concluding that it started 168th in the race. In feet, however, entries for Paris-Madrid totalled 814, and although 39 of these had been cancelled by the time that entries closed, every entrant was apparently left with his original number on the entry list : I do not know who had the number 314, but Gibed actually started in the race carrying the number. 313 on his 24-h.p. Gillet-Forest.

The 275 entries which remained consisted of 112 heavy ears, 64 light cars, 40 voiturettes and 59 cycles, but of these a fair number proved to he non-starters. Rose, in fact, shows that there were actually at the start 90 heavy ears, 49 light cars and 36 voiturettes, making a total of 175 cars of various kinds and an unspecified number of cycles, of which 15 finished. The total number of starters must, therefore, have lain between a minimum of 190 and a maximum of 234, the actual figure being, I believe, about 225. Already, therefore, it is apparent that a good many numbers were missing from the list of starters. The matter is;, however, further coin

plicated by t fact IIi t wl mile the heavy cars and the ligitt cars were started indiaeriminately, :Lecordilig I u Bair numbers, all of thciit wen sent off before any of the presumably slower voiturettes and cycles were despatched. Any of these latter, therefore, which parried lower numbers than Gabriel's Mors were nevertheless despatched after the winner.

Having thus cleared the ground to sonic extent it now remains to see how many ears did in fact start before the Mors, which can be determined by seeing how many of the heavy curs and light cars in Rose's list had lower numbers ; and after several recounts I have come to the conclusion that the answer is 81. Gabriel, therefore, started not 168th but 82nd, and the mystery of Paris-Madrid begins to clarify somewhat. We can now cheek this result to some extent by seeing whether, if he started eighty-second, Gabriel arrived in Bor ? lemix at :khottl the t lila, one would expect him. iii I 44,! " Reeord " we only have, of course, the nci racing 11cctes taken by the cars for t he joarney, and to this must be added the they spent in the " controls.' Following an investigation which Mr. Rose has recently made in Paris, and the results of which he has kindly communicated to me, we know that the standard addition to be made to the net tithes on this account is 2 hours 54 minutes. The first car to leave Versailles, where the nice actually started, was a 45-h.p. de Dietrich driven by Charles Jarrott, who should have got away at 3.80 a.m.; but as at that unearthly hour it was considered to be still too dark, his start was in fact delayed until 3.45 a.m. The first Clit to reach Bordeaux, at 12.15 p.m., was Louis Renault's 80-h.p. light car, which started third, that is to say at 3.47 a.m. Renault's time was 5 hours 29 minutes 39.2 seconds, to which must be added control additions, in this ease, Of 2 hours 57 Minutes, and 12.14 p.m. is the actual time he should have arrived at Bordeaux. "Sixteen minutes behind came Jarrott," says the "Record "; that is to say he arrived at 12.3) p.m., and if in this case one adds control additions of 2 hours 53 minutes to his running time of 5 hours 52 minutes 55 seconds, the calculation again cheeks up

"After Jarrott had disappeared into the town," continues the • Record," "there was a long wait. Nearly forty minutes slipped by, and then Gabriel drove in." Unfortunately : this does not give us a very precise time for Gabriel's arrival. If it means that this took place nearly forty minutes after Jarrett's, then We should expect Gabriel at about ten past one. If, on the otherhand, it means nearly forty minutes after Jarrott had disappeared into the town, then obviously Gabriel arrived a few minutes later than this. Starting eighty-second, he would have got away from Versailles at 5.00 a.m.; his running time was 5 hours 14 minutes 81.2 seconds' the control Addition in this case is 2 hours 53 minutes ; and we should therefore expect Gabriel at Bordeaux at 1.13 p.m. This is so close to the time at 'which he evidently did arrive that we can I think, rest quite content with the eafeidation that he did in fact start eighty-second.

Of the 81 cars which started before his, only two reached Bordeaux before Gabriel's, and somehow, therefore, the winner must have got past the other 79 on the road. It is not necessary, however, to suppose that he passed all of them at Speed, and in a cloud of dust. As a matter of fact, of the 175 cars which started in Paris-Madrid, only 99 reached Bordeaux, and it would thus be statistically possible for Gabriel only to have passed three moving cars on the road! In actual fact, of course. all 70 which failed to finish did not start before the winner, and, of the 81 which did start before him, the number which fell out was 37. Unfortunately, in the majority of cases, we do not know whether they retired before Gabriel caught up with them or afterwards. Of the remaining 42 cars, all of which Gabriel somehow passed, and all of which reached Bordeaux, a proportion, no doubt, had temporary stops, and Gabriel may have passed -some of them while they were changing tyres or effecting repairs at the roadside; but even allowing for this uncertainty, it would seem reasonable to estimate that he got by something between forty and sixty (taxa while they were on the move. We can, moreover, glean a little more information as to what he did and when. By the time Ranibouillet was reached, we learn from the " 'Record," " Jenatzy had, passed sonic sixteen cars and Gabriel about twenty-five." This is really extraOrdinary, because the racing distance to. Rambouillet was only 17i miles. Of the twenty-five ears which, started inunediately in front of Gabriel, eleven failed. to reach Bordeaux and some of them. may have dropped out right at the start.. Two more made very slow total times, and