434 



SCIENCE. 



[Vol. XIII. No. 331 



practicable, any mixture of water and spirit, which would tend to 

 vitiate the experiments. 



The heat from the gas-flame is first taken up by the water in the 

 boiler, and is then passed on to the copper coil, and evaporates the 

 spirit ; the water only acting as a convenient means of transmitting 

 the heat from the flame to the spirit. By these means it was 

 possible to try steam and spirit vapor under precisely similar cir- 

 cumstances with regard to boiler efficiency. 



The experiments consisted of several continuous trials, each of 

 three hours duration, alternately with steam and with spirit vapor. 

 The upshot of these experiments was, that although the amount of 

 gas consumed during the three hours was practically the same, 

 being at the rate of 82 and 83 cubic feet, the power obtained, as 

 tested on the brakes, was, in the case of spirit, nearly twice that 

 recorded in the case of water, the powers being as 4,722 to 2,524. 



At equal intervals during these trials, diagrams were taken with 

 an ordinary indicator. The working-out of these diagrams gives a 

 power, in the case of spirit, of 1 1,975 foot-pounds per minute, and, in 

 the case of water, of 5,199 foot-pounds per minute, which more 

 than confirms the results obtained by the brake. 



Touching the class of spirit used for the experiments, Mr. Yar- 

 row mentioned that it was a hydrocarbon distilled from petroleum, 

 having a specific gravity of about .680. The reason this spirit was 

 adopted was because it is low in price, and can be easily procured ; 

 and also, being obtained from petroleum, it is of an oily nature. A 

 spirit which is not oily in its character would be deficient in lubri- 

 cating-power, and therefore not so suitable for working the ma- 

 chinery. 



Touching the evaporation of spirit to produce power in actual 

 practice, at present the only application which has been success- 

 fully developed is for the propulsion of launches. It is termed the 

 "Zephyr" system. For small sizes, certainly there would seem to 

 be no question, that, where the spirit is obtainable, this system is 

 destined to take the place of steam, not so much on account of the 

 probable increased efficiency, as the general convenience of the 

 arrangement. A launch propelled on this plan, 36 feet long by 6 

 feet beam, and built of steel, has a total weight, including machine- 

 ry, of only one ton ; while, had it been propelled by steam, the 

 weight would have been considerably greater. 



Mr. Yarrow did not venture to assign any reason for the appar- 

 ent gain in the use of spirit over water, but pointed out, that, in a 

 condensing-engine, the two great losses of heat are due, first, to 

 the waste gases which pass up the funnel, and simply go to raise 

 the temperature of the surrounding air ; and, second, to the loss 

 arising through raising the temperature of the condensing-water, 

 which goes to warm the sea. 



As regards the first loss, it is self-evident, that, owing to the low 

 temperature at which the spirit evaporates, the products of com- 

 bustion are available to produce evaporation down to a much lower 

 temperature than in the case of water. As an illustration of the 

 low temperature at which the waste gases pass away in the 

 " Zephyr " launches, when going at full speed, it is quite possible to 

 hold one's hand immediately over the funnel. 



The results of these experiments give reasonable ground for be- 

 lieving that this system, in some form, is capable of further and 

 possibly extensive development. 



WEST INDIAN HURRICANES. 

 June is the first of the five months that constitute the hurricane 

 season, and every navigator bound for the tropics should now be on 

 his guard, and watch for the earliest indications of an approaching 

 cyclone, lest it may be of hurricane violence. The " Pilot Chart " 

 last month contained a small chart illustrating the general tracks 

 along which these storms move, together with a diagram explain- 

 ing the importance of the fact that the tracks recurve in a latitude 

 dependent upon the month. In June, for instance, the point of re- 

 curvature is in about latitude 20** to 23° north, and here the entire 

 storm moves in a northerly direction, while in lower latitudes it 

 moves north-westerly, and in higher latitudes north-easterly. Each 

 of the accompanying diagrams is therefore applicable to a slightly 

 different belt of latitude, according to the month, as stated there- 

 with. There may be an occasional exception, but a navigator can 

 at first only be guided by the general rule. 



A careful study of these diagrams will, it is thought, enable a 

 navigator to form a clear idea as to the general character of West 

 Indian hurricanes. It should, of course, be understood that no 

 two hurricanes are exactly alike ; and a master of a vessel can only 

 be guided by what is known from the experience of others, his own 

 experience, and the indications that he can obtain from the weather, 

 wind, barometer, cloud-movements, etc. It is now generally rec- 

 ognized that the old 8-point rule is not always a safe guide for 

 action ; and, although the storm-card based upon it has been pub- 

 lished on the "Pilot Chart" up to the present time, efforts have been 

 made to explain the spirally inblowing character of the winds in a 

 cyclone. The old storm-card and the 8-point rule can no longer 

 be recommended, except, perhaps, as a rough-and-ready rule for 

 those who will not take advantage of the progress that has been 

 made since this old rule was established. By looking at any one 

 of these diagrams, for instance, it will be seen that the bearing 

 of the centre is hardly ever exactly eight points to the right of the 

 wind, but is generally considerably more than that, especially in rear 

 of the storm. It will be noticed, also, that the storm is never 

 exactly circular, and that its shape changes as it moves along its 

 track. The limited space on the chart forbids any detailed dis- 

 cussion of the subject'; but the following brief resume, prepared for 

 the information and guidance of practical navigators during the 

 present hurricane season, is taken from the June " Pilot Chart : " — 

 The hurricane regions embrace the tropics north of the loth 

 parallel, the Caribbean Sea, Gulf of Mexico, and a broad belt curv- 

 ing north-westward from about St. Thomas, and following the 

 Gulf Stream toward the Grand Banks of Newfoundland. 



The earliest indications are. unusually high barometer, with cool, 

 dry, fresh winds, and very transparent atmosphere ; a long, low 

 ocean-swell from the direction of the distant storm ; light, feathery 

 plumes of cirrus clouds, radiating from a point on the horizon 

 where a whitish arc indicates the bearing of the centre. 



Unmistakable signs of a hurricane are the following : As the 

 cirrus veil spreads overhead, with halos about the sun and moon, 

 the barometer begins to fall, slowly but steadily, and the ocean- 

 swell increases ; the air becomes heavy, hot, and moist ; dark red 

 and violet tints are seen at dawn and twilight ; the heavy cloud- 

 bank of the hurricane soon appears on the horizon, like a distant 

 mountain range ; the barometer falls more rapidly, and the wind 

 freshens, with occasional squalls of fine, misty rain. 



As to general size and velocity of progression, the storm area is 

 smaller in the tropics than farther north, the cloud-ring averaging 

 about 500 miles in diameter, and the region of stormy winds 300 

 miles or even less. In low latitudes the entire storm moves west- 

 ward and north-westward about 17 miles an hour ; in middle lati- 

 tudes, north-westward and northward, moving more slowly as it 

 recurves ; and finally north-eastward, with a velocity of translation 

 of 20 or even 30 miles an hour, its area increasing rapidly as it 

 follows the Gulf Stream toward the Grand Banks, and sweeps 

 across the Atlantic toward northern Europe. 



One of the most important indications that an approaching storm 

 is of hurricane violence is the marked cyclonic circulation of the 

 wind, lower and upper clouds, etc. This may be easily appreciated 

 by remembering that a cyclone of any great intensity is an ascend- 

 ing spiral whirl, with a rotary motion (in the northern hemisphere) 

 against the hands of a watch, as shown on the diagrams. The 

 surface wind, therefore, blows spirally inward {not circularly, ex- 

 cept very near the centre) ; the next upper current (carrying the 

 low scud and rain clouds), in almost an exact circle about the 

 centre ; the next higher current (the high cumulus), in an outward 

 spiral ; and so on, up to the highest cirrus clouds, which radiate 

 directly outward. The angle of divergence between the successive 

 currents is almost exactly two points of the compass. Ordinarily, 

 with a surface wind from the north, for instance, the low clouds 

 come from the north also ; on the edge of a hurricane, however, 

 they come froin north-north-east invariably. In rear of a hur- 

 ricane, the wind blows still more nearly inward. With a south- 

 east wind, for instance, the centre will bear about west, the low 

 clouds coming from south south-east (two points to the right of the 

 wind), etc. Great activity of movement of the upper clouds, while 

 the storm is still distant, indicates that the hurricane is of great 

 violence. 



