From this table it appears, first, that only five woods-the Chella woonjay, Eloopay, Kurrah murdah, Palay, and Karavalum—were sound at the end of two years, and that the other eleven species would not last as sleepers for that time. Secondly, it is evident, that where the sleepers were uncovered, decay was much less rapid than where they were completely buried in the ballast. The last two woods in the list were not found in the two systems of ballasting; of the remaining fourteen it will be seen, that in ten the advantage is decidedly in favour of the uncovered sleepers, while in the other four, the difference in favour of the covered sleepers is extremely small.

The system of leaving the sleepers as much as possible uncovered has many advantages. It effects a saving of about 9 cubic yards of ballast per chain of line, and there is proportionately less labour in opening out the road for examination, or repairs. It allows the platelayers to detect at once any defects in the substructure of the road, which, when the sleepers are buried in ballast, can seldom be done, until the effects are visible in a depression of the rails; and by throwing off wet more rapidly, it keeps the sleepers drier, and induces their better preservation.

The objections urged against the system were, that the heat of the sun would split the sleepers, and cause the keys and trenails to shrink. Further, that the road would not be sufficiently bedded

in the ballast to prevent its twisting when the rails expanded, or on the passage of trains.

Regarding the splitting of the sleepers, the following statement, referring to the whole of the sleepers examined, shows the effects of the two systems:

It thus seems that, while amongst the uncovered_sleepers there is a larger proportion beginning to split' or 'useless from being split,' there is, on the other hand, amongst the covered sleepers, a still larger proportion beginning to rot,' or 'useless from being rotten; and the per centage of good sleepers amongst the covered is less than amongst the uncovered. On the whole, therefore, it appears to be decidedly better not to bury the sleepers in the ballast, as has generally been done. As regards the trenails, the Author found that where the sleepers were rotten the trenails were almost invariably rotten too, on which account uncovered sleepers seem to be preferable. On the other hand, he never found that trenails whose heads were exposed to the sun were loosened by shrinking. Keys, if not properly seasoned, will shrink in India. No system of ballasting which would protect them from the sun would be admissible, as it would effectually prevent the platelayers from tightening them up. Besides this, the Author found, that where the ballast touched the keys, whether of English, or of native wood, they were generally soft, and covered with fungus, and, where the line was not opened for traffic, in many instances they were eaten by white ants. With reference to the lastnamed objection, that the road would be more liable to twist and buckle if the sleepers were not buried in the ballast, the Author can only state, that he never knew this to be the case on 40 miles of line thus ballasted and opened for traffic, on the division under his charge.

Trenails made in India cost from £2 108. to £4 per thousand; the price varying with the difficulty of getting wood and labour. The woods generally used for the purpose were the Vengay, Kurrah murdah, Erool, Porasa or Satin wood, and Trincomalee. These trenails appeared excellent, but the Author left India before he had an opportunity of observing their durability. He is, however, convinced, that they are equal to any trenails sent from England. Trincomalee wood in particular seems deserving of more attention

for the manufacture of trenails. Keys were also extensively made in India, of the Vengay, Kurrah murdah, Erool, and Teak. These keys were considerably longer than usual, so as to allow of their being driven up, when they got loose. Teak keys seemed to be the best, and as their cost did not exceed £6 per thousand, it would be worth while to have them supplied in India. The other woods named, if properly seasoned, would doubtless be found as durable as, and cost less than, Teak.

From personal observation and inquiry, the Author thinks the following twelve woods will make really good sleepers, to be used plain :

The previous list contains the names of eleven woods which should, in the Author's opinion, be rejected as unfit for sleepers, if used in a plain state.

It is deserving of notice, that of the sleepers beginning to rot' 19 per cent. of the whole had begun under one, or both chairs. In many of the sleepers which were so rotten as to be useless, it was impossible, from their state of decay, to determine where, or how the decay had commenced. Speaking, therefore, only of those previously alluded to as 'beginning to rot,' the additional fact was observed, that in the uncovered sleepers 24 per cent., and in those covered 18 per cent., decay had begun under the chairs. The Author believes that the cause of this decay is the retention of moisture under the chairs, in the hollows of the castings. The rarefaction of the air within this hollow would be greater when the chair was not protected by ballast from the heat of the sun: when rain fell, this air would be sooner cooled, and the water would have freer access to the interior, than where ballast intervened. It may be said, that under any circumstances the quantity of water so retained must be extremely small. Nevertheless for many hours after the rest of the sleeper was perfectly dry, damp was observed all round each chair, the water within being slowly expelled, as the chair was again heated. The large proportion of sleepers shown to have been affected by this damp, and the fact that many sleepers had to be removed, because they were rotten

under the chairs, while the rest of the sleeper seemed perfectly good, are sufficient proofs that this subject deserves consideration. It is believed that if the seats of the chairs in the sleepers were well tarred, it would go far to remedy this evil; although the Author had not an opportunity of witnessing the effects of what he recommends. It cannot be said that the chairs had, in any instance, been crushed into the sleepers, for at the date of the second and last annual examination, nothing heavier than a platelayer's 'trolly' had passed over them.

With regard to the second cause of failure,-the woods being used too young and unseasoned,-the list already given, of the state of the sleepers at the end of the second year, does not show the actual expenditure of sleepers on the Madras Railway up to that time. Large as the renewal is there seen to be, it was even yet larger in reality. For as the main object of the investigation was to discover what woods were good and what bad, so in the record made of the results of this investigation no notice was taken of the decay of sap wood, which unfortunately formed a considerable portion of many sleepers; its decay, which took place always before the end of the first year, leaving so little of the original sleeper as to necessitate its renewal. In such cases when the heart wood was found to be sound and good, the sleeper was recorded under that head.

6

The circumstances under which the sleepers were supplied were such as to account, in some measure, for the great loss from this cause. In the first place, until sleepers were actually required for the railway, there had not been any demand for them from any quarter. The jungles were unknown to all but a few native wood-cutters. A commencement was made by some of the civil servants of the Government, who rendered valuable assistance to the Railway Company, by working the jungles in their neighbourhood. The example thus set was followed by natives possessed of capital, who entered into contracts with the Railway Company for the supply of sleepers. These men sublet the work to. wudders,' a peculiar tribe accustomed to wood-cutting in the jungles. But this arrangement was speedily found to fail. The native contractors took little interest in the work, and exercised little supervision over the wudders. They secured themselves by throwing upon the wudders, a poor ignorant race of men, all the loss caused by sleepers being rejected by the Railway Company's agents. The want at that time of any supervision in the forests, (now ably exercised by the Forest Department), tempted these wudders, and the rude appliances at their command forced them, to leave untouched large trees from which really good sleepers might have been cut, and to satisfy themselves with felling only the smaller and young trees. They also frequently lopped off the branches of large trees, and left fine

trunks standing. Of course a large proportion was rejected by the Engineers of the Railway Company, but the supply was so scanty and irregular, that at last, when the permanent way was being laid, sleepers had to be obtained by any means, and many sleepers which would have been rejected at first were bought up eagerly at last. In fact instances occurred, in which sleepers rejected by one Engineer, who could afford to pick and choose, were taken immediately to a less fortunate neighbour, (the marks of rejection branded on them being carefully removed), and by the latter they were bought and used. Under these circumstances it is not to be wondered at, that so many inferior sleepers were used. The appointment of Conservator of Forests was made in December 1856, but some time elapsed before the arrangements could be completed for an efficient supervision and systematic working of the forests. It was not, for instance, until July 1858, that an establishment was sanctioned for the working of the Salem Jungles, from which a large portion of the railway was supplied with sleepers. Things continued in this state until May 1858, when the system of giving large contracts for sleepers was discontinued, on account of the utter inability of the contractors to fulfil their engagements, and the Engineers received permission to purchase sleepers direct from any one who should offer them for sale. This change had many advantages. The middle men being done away with, the Engineers had direct access to the wood-cutters. In many cases the price of the sleepers was reduced, but where it remained the same, the wood-cutters who brought the sleepers direct to the Railway Company received a larger profit, than when they had to work for contractors who were not remarkable for liberality, nor even for honesty in their dealings with the poor, ignorant labourers. But many of the disadvantages of the old system still continued. The different establishments of the Forest Department were at first unable to superintend and control the numerous parties engaged in all directions in felling timber, and the wood-cutters were still no further advanced in the knowledge of the good woods, or the means of obtaining proper appliances for felling any but the smallest trees. The results were, therefore, although in a less degree, the same as before. Sleepers were used green and unseasoned. The supply of good sleepers was far short of the demand, and timber of inferior quality had to be bought at last, to complete the laying of the permanent way.

From what has been said, it will be seen that the sleepers which have failed on the Madras Railway may be divided into two classes-those which are of perishable woods, and those which, although of good woods, have been cut from young trees. The existence of the first is to be attributed to want of experience of the nature of the Indian woods; and of the second to want of a