During recent years many attempts have been made to increase the velocity of projectiles and at the same time to keep down the maximum pressure of the powder gas within the bore of the gun. It is evident that if the powder pressure is to be restricted, an increased velocity can only be obtained by causing the .maximum pressure to act during a longer period, and over a greater space, on the projectile, and with this object the various slow-burning and cocoa powders have been produced, and tried with results which, though fairly successful so far as keeping down the pressure is concerned, can hardly be considered satisfactory, as the large amount of hot residuum left in the bore after firing has given rise to serious difficulties in reloading the guns, as well as causing danger of premature ignition on reloading.
Thus it is stated in "The extracts from the Proceedings of the Department of the Director of Artillery and the Ordnance Committee," that:
1. In 8-inch Armstrong B. L. gun, in which internal metallic tubes in cartridges were used: "The presence of molten residue in the chamber when the breech was opened was very apparent on more than one occasion."
2. In 12-inch B. L. gun, Mark II, in which Rottwell Prism' powder was used: "At the commencement of the practice a couple of buckets of water were thrown into the gun after each round, to soften the fouling."
3. In 12-inch B. L. gun, Mark I, in which Cocoa powder was used: "Considerable delay was experienced owing to an accumulation of fouling in the chamber, and the shells had repeatedly to be tapped home."
4. In 12-inch B. L. gun, Mark I, in which Brown Prism' powder was used: "A scraper was used throughout for removing the residue."
5. Residuum from Rottwell and Cocoa powders: "With the Rottwell and Cocoa powders a large quantity of residue was left; it was so hot at first that it actually fired prisms of powder even when enclosed in a shalloon cover."
Nor are these Cocoa or Brown powders altogether satisfactory as regards uniformity of pressure, for we find from the above named "Extracts" that the Superintendent of Royal Gunpowder Factory states:
6. "He cannot account for the abnormal results which occur sometimes with all powders; for instance, it will be admitted that Rottwell Cocoa is a good powder, yet with the same projectile and the same sample of powder fired the same month, July, 1888, the results in the I0.4-inch B. L. gun were:
Charge (lbs) Muzzle velocity (foot-secs) Pressure (tons)
260 2213 22.35
240 2149 24.0
where the lower charge gave lower velocity but much higher pressure."
7. That the Superintendent of Royal Gunpowder Factory states also: "The results of experimental powders compared with 132, Prism, and Cocoa, from which it will be seen that Cocoa powder has given very different results at different times—there being as much as 63 feet muzzle velocity and 3 tons pressure between different rounds with same charge and projectile—there is, apparently, a definite superiority over the best black powder."
(This remark refers only to black powder as usually made at the present time.)
Furthermore, these powders have been found to be very erosive in their action, this erosive action being due to the intense heat generated by them, as well as to the chemical constitution of the products of combustion. And in addition to this, it is still somewhat doubtful whether they do not deteriorate when kept for any considerable period in extreme climates. On the other hand, with the black powder, the fouling of the weapon after firing is much less than with any variety of slow-burning Brown or Cocoa powder yet produced, and it will perhaps be admitted that any method of rendering this black powder, the good qualities of which have been so long known, capable of giving high velocities with low pressures will be worthy of exhaustive practical trial.
At the same time it must be understood that the Quick method of forming the powder, or more properly "propelling material," is equally applicable to any of the new chemical combinations for propelling materials as to the ancient, saltpetre, charcoal and sulphur mixture.
Before entering into a description of the way in which this end has been attained in the Quick perforated cake powder, it may be advisable to take up some general considerations concerning the combustion of powder in the chamber of a gun.
It is obvious that such a value would be difficult to obtain in practice, yet the more nearly it is approached the greater will be the velocity of the shot with a given maximum pressure, that is, provided the assumption that most of the work is done during expansion be fulfilled.
There remain two methods of increasing the value of v„, the first being to increase the air-spacing of the charge by increasing the volume of the chamber, and the other is to make use of a slow burning powder, which causes the projectile to move a sensible distance whilst combustion is proceeding, and it is in this direction that the best results are to be looked for.
The rate at which powder burns away is dependent upon a number of conditions. It varies, first, with the nature of the material from which, and the temperature at which, the charcoal contained in it has been prepared; second, with the amount of grinding or incorporation its constituents have undergone; third, with the amount of moisture it contains, and with the length of time and temperature of storing; fourth, with the density of the powder itself; and fifth, with the pressure of the atmosphere of fire to which it is exposed.
Thus comparing No. r gun with No. 4 gun, we see that we may
use 10 per cent (or -exactly 9.72 per cent) more pressure in No. 4 gun than in No. 1 gun; whilst as regards weight per inch length of chamber, the No. 4 gun will be 6.3 per cent less in weight (or No. gun is 6.92 per cent heavier than No. 4 gun), although the thickness of the wall of the chamber is the same in No. 4 gun as in No. 1 gun. Or if we make a gun of 21 inches diameter we have a gun of 0.787 ton greater strength than No. 1 gun, and 15.7 per cent less weight (or No. 1 gun is 18.6 per cent heavier than No. 5 gun) per inch of chamber length.
Of course, with the chambers of small diameter a greater length of chamber would be required to contain the same charge of powder with the same density of loading, that is, with the same amount of air-spacing, and greater length of chamber would add to the weight of the gun somewhat. But, as the gun with small chamber would be stronger, a higher gravimetric density may be employed, so that the total volume of the chamber would be less, provided that the charge be properly ignited in the first place, and the powder converted at a uniform or increasing rate into gas. The central air-spacing and the form of the cakes of the "Quick" powder insure this necessary instantaneous ignition and complete combustion of the whole charge at a uniform or increasing rate.
With regard to the maximum pressure to be adopted for the powder gas in the gun, this is a question to be settled by practical experience rather than by speculative deductions. It would appear that a pressure of 20 tons per square inch should not be exceeded, and one considerably lower will probably be found to give the best results. If higher pressures are employed, there appears to be great danger of reaching, at the high temperature at which the powder burns, the temperature of fluidity of the metal of the bore, in this way increasing erosion. It is well known that iron and steel at high temperatures are porous to many gases, and it is not unlikely but that with high pressures and very high temperatures combined, a not inappreciable quantity of the gases produced may be forced into the metal, so that there would occur a larger loss of energy from the cooling action of the surface of the chamber and bore, with increased heating and possible deformation of the gun, than with gases of lower pressure and less temperature. It seems probable that some action of this kind must occur, as it is otherwise difficult to account for the observed facts that high pressures are frequently associated with low velocities. In addition to this loss of energy, this penetrative action of the gases will effect a change in the molecular condition of the metal, by which it will be deteriorated, which deterioration will tend to increase erosion as firing proceeds. From these conditions it would appear that the maximum pressure to be admitted is not merely a simple question of thermodynamics, as is sometimes considered, but is limited by the physical condition of the metal of the gun, besides the mere tensile strength thereof.
It is but an evident truism to say that the more work the powder does on the gun, the less work can it do on the projectile, and conversely, that the more work that is done on the projectile, the less work and the less wear, tear and injury will be done on the gun.
Here it may be remarked, that by constructing the powder for heavy guns on the principles stated above, the rise of pressure in the bore can to a very large extent be effectively controlled by suitably arranging the initial ignition surface, the density of the cake-powder, the air-spacing, and the form of the cakes, so as to give a nearly uniform or increasing combustion surface. It would therefore appear that such a powder is peculiarly suited to firing large projectiles of thin metal charged with high explosives, and the final solution of a satisfactory dynamite gun is certainly to be found in this direction, as the pneumatic type is open to many grave practical objections as to its length, the cumbrousness of the machinery required for manipulating it, and the want of accuracy in such high-angle fire.
The conclusions which have been arrived at above may be summarized as follows:
1. The pressure should, if possible, be maintained at its maximum till the projectile has traveled some considerable distance up the bore.
2. This can be most satisfactorily obtained by using a powder whose surface of ignition increases as combustion proceeds, or remains practically uniform.
3. That a comparatively low maximum pressure should be adopted, if the best results, in the fullest sense of the term, are to be obtained.
4. That the density of the powder should be very high, so as to secure uniformity of combustion, and that the initial ignition and terminal combustion surfaces should be large enough to secure a sufficiently rapid combustion of this very dense and slow-burning powder, so that none may be blown out of the gun in an unconsumed state.
5. That the air-spacing should be reduced to the lowest possible amount compatible with the due ignition of the charge, and with a sufficient channel space to allow of a fair flow of the powder-gas, direct to the base of the projectile as soon as the charge is ignited.
6. That the powder-charge should be so arranged that the whole of it should be simultaneously ignited, in order that each and every element (cake) should be surrounded by an atmosphere of fire equal in pressure in all directions.
It is desirable to consider some practical application of the principles before stated, and to give some examples of the manner in which different descriptions of powder burn, and some results which have been obtained with the perforated cake powder in the Quick B. L. gun of 3-inch bore.
The following is a statement of the area of the combustion surface of 42-pound charges for a 6-inch gun, the charges being of various powders:
a. 42-pound charge of 1.5-inch cubes of powder, averaging 6 to the pound, P2 powder, density 1.75.
b. 42-pound charge of hexagon powder, .976 inch high, 1.367 inch over sides, with t-inch central perforation, .394 inch diameter, prism 2 powder, density 1.75.
c.. 42-pound charge of Quick perforated cake powder, 6.7 inches diameter, thickness or length on axis 1.7 inch, with central perforation 2.5 inches diameter, and 6o other perforations .25 inch diameter, density 1.75.
d. 42-pound charge of Quick perforated cake powder, 7.7 inches diameter, thickness or length on axis 1.75 inch, with central perforation 1 inch diameter, and 72 other perforations .3 inch diameter, density 1.75.
e. 42-pound charge of Quick perforated cake powder, 7.7 inches diameter, thickness or length on axis 1.75 inch, with central perforations 2.5 inches diameter, and 33 other perforations .25 inch diameter, density 1.80.
Then assuming that all these powders are of uniform quality and density, except charge e, their rate of combustion under any given pressure will be the same, and we shall find that in the case of a the initial ignition surface will have an area of about 3400 sq. inches. When half the weight of the charge has been consumed (that is, converted into gas), the surface of combustion will be only 2130 sq. inches in area, and this area will diminish until zero is reached. Thus (1) at the instant when the projectile is without motion and tightly sealing the bore of the gun, the gas-generating surface of the powder is at a maximum. (2) When half the powder has been consumed and the projectile has acquired a certain velocity and has moved a considerable distance up the bore, the gas-generating surface is reduced about 38 per cent. (3) The gas-generating surface will rapidly diminish after this; and as the pressure of the atmosphere of fire in the bore falls also as the shot travels towards the muzzle, so will the velocity of combustion due to the pressure of the atmosphere of fire also decrease, so that it is highly probable that some of the powder will be blown out unburnt.
If we now consider the time of combustion of this form of powder, and assume that a thickness of .4 inch of solid powder will be burnt in 1 second of time under atmospheric pressure, then we shall have for the time during which this 1.5-inch cube is burning 1.875 second. (The actual velocity of combustion at the atmospheric pressure is about .4 inch per second; but at the usual pressure of the atmosphere of fire in large guns, M. Castan estimates that the mean velocity of combustion is about 32 times greater, that is, 12.8 inches per second.) The law assumed by de St. Robert, Castan, and Sarrau, that the velocity of combustion of powder varies as the square root of the pressure of the atmosphere of fire, may be true of very dense hard powders, and may be altogether inaccurate as regards powder of less density and greater porosity. It is highly probable that the velocity of combustion of porous powders of low density will be very much greater, owing to the flame being driven into the pores of the powder by the higher pressures of the atmosphere of fire. Again, the law assumed by Piobert, that the velocity of combustion varies inversely as the square root of the density of the powder when burnt under the same pressure, may be true when the powder is burnt under atmospheric pressure only, but may be totally untrue when burnt under higher pressure; the velocity of combustion rising much more rapidly than inversely as the square root of the density when low density powders are burnt under high pressures.
Then, although every cube in the whole charge may have been ignited simultaneously, we shall find that when half the time of combustion has expired, the area of combustion-surface will be only 850 square inches, the weight of powder burnt during the first half of the period of combustion being 37.29 pounds, and in the last half of the period only 4.71 pounds of powder remain to be burnt.
But we know that in charges of this kind of powder, simultaneous ignition does not take place, as the rear of the charge is first ignited.
Repeating these calculations for the other charges of powder and placing them in tabular form (as on the next page), we can compare the results; the velocity of combustion being calculated for the pressure of the atmosphere of fire in which the combustion is effected.
From these results we can see that the initial ignition-surface of the Quick cake powder c is about 24 per cent less than that of a and 35.5 per cent less than that of b, and that the gas-generating surface of the Quick cake powder is at a maximum when the resistance of the projectile is at a maximum. Also that whereas the gas-generating surface, and consequently the gas generation, falls rapidly in the charges a and b, in the case of c and d the gas generation steadily increases as the projectile rushes up the bore.
We must then pursue our investigation as to the time of combustion of the powder, and ascertain the amount of powder which will have been burnt at the instant when half the time of combustion has expired, assuming as at page 420 that 0.4 inch of powder is burnt through at atmospheric pressure in I second, that is, that a 1.5 inch would occupy 1.875 second in burning.
From these results we find that with the Quick cake powder we can obtain a progressive powder although the density may be very high and uniform throughout, and the quality uniform. And it is submitted that this form of powder possesses great advantages over the Fossano and other so-called "progressive" powders. It is also beyond question that a larger amount of energy has been obtained per pound of black powder than has been obtained per pound of any other cannon powder, when fired under the same conditions. And by the large reduction of air-spacing which the Quick form of powder permits, a very much higher efficiency will be obtained than has ever been the case hitherto.
Thus, in the record of experiments with the Quick experimental 3-inch B. L. gun, we find in round 29, where the air-spacing was small, due to the size of the chamber (145 cubic inches), giving 26 cubic inches per pound of powder, the efficiency of the powder was 71 foot-tons; whereas, in round 79, after the volume of the chamber had been increased to 18o cubic inches—giving 36 cubic inches per pound of powder—the efficiency of this 5-pound charge of black cake powder was only 56.6 foot-tons per pound of powder.
Now an absolutely perfect propelling agent (which of course is only imaginary) would be a material which would occupy say one-thousandth part, or less, of the length of the bore of the gun (no chamber being required), which material should give off such a volume of permanent gas (without color, smell, or active chemical properties) at such a rate as would give a uniform pressure of, say, 15 to 20 or 30 tons per square inch throughout the whole length of the bore of the gun, the temperature of the said gas not exceeding, say, 500° to 1000° Fah. Such a propelling material we certainly have not at present, nor is there any prospect of its being obtained.
Such a propelling material would then require that the guns should be of the same strength from breech to muzzle—that is, they must be cylinders of the same diameter throughout—which would give the lightest and shortest ideal gun for obtaining any given effect.
Now, although we have not this ideal propelling material, we have the black powder material which, so far, has been proved to give a greater amount of ballistic energy per pound weight than any of the brown materials, when fired under the same conditions. (The disadvantage of the black powder material has been that it has given higher pressures than the brown material, but we need not refer further to that point at present.)
Now, by the perforated cake formation of the black material it has been proved that we get these advantages:
1st. That the complete ignition of the whole charge, however long, is effected simultaneously.
2d. That we can get either a uniform or an increasing area of combustion, and consequently a very uniform pressure during the whole time of combustion.
3d. That the pressure and ballistic effects of a given weight of this black powder are far higher than a similar weight of brown cake powder will give under the same conditions.
For from the experiments of June, 1887, with the Quick 3-inch B. L. gun, in round 28 we obtained an observed velocity of 2000 feet with 12-pound projectile and 5.56 pounds of black cake powder of 1.77 density, and a pressure which did not exceed 14 tons per square inch. In round 29 we obtained an observed velocity of 2145 feet , with 12-pound projectile and 5.51 pounds of black powder of density 1.75, with a pressure of 14.05 tons. In round 41 we had an observed velocity of 2011 feet with 1 2-pound projectile and 6 pounds of black cake powder of 1.71 density, and a pressure of 16.8 tons per square inch, which, comparing with the foregoing rounds, shows that this lighter, quicker-burning powder, whilst giving a higher pressure, gave a less velocity than the smaller charge of higher density powder used in round 29. This result indicates that the density of the powder was too little for the weight and resistance of the shot and for the size and strength of the gun. In round 28, the density being higher again than in round 29, the pressure was slightly lower and the velocity considerably less, showing that the density of the powder was too high for the weight and resistance of the shot and for the length of the gun, the efficiency of the powder per pound being 63.4 foot-tons. The round 29 shows that the density of the powder gave such a rate of combustion as to give probably the maximum results for the particular weight and resistance of the shot and for the length of the gun, the efficiency of the powder being 71 foot-tons per pound of powder. Turning now to round 78, we find that 5.25 pounds of brown cake powder of density I.8o to 1.83 gave an observed velocity of only 1435 feet to a projectile weighing 12.25 pounds, the cakes being 1.5 inch thick. Thus the efficiency of the brown powder was only 32.8 foot-tons per pound of powder. (This charge consisted of 7 brown and black cake. If the black cake is credited with the same efficiency as that used in round 79, then the actual efficiency of the brown powder will be found to be only 28.1 foot-tons per pound.) Round 79 was fired immediately after round 78. In this round No. 79, 5 pounds of very light density powder were fired with 12-pound projectile, the cakes being 1.5 inch thick, as in round 79. The observed velocity was 1845 feet, giving an efficiency of 56.6 tons per pound of powder. (See page 422 respecting reduction of efficiency due to increased air-spacing.) The air-spacing and the other conditions being in all respects exactly the same as in round 78, we find that the black powder was 72 per cent more effective than the brown powder.
Now the larger and the longer the gun is, the greater the weight and resistance of the projectile, the more dense may be the powder cakes, providing the material will burn with sufficient rapidity to be converted into gas before the projectile has accomplished more than 43 per cent of its travel in the gun, because the projectile is a longer time traveling up the bore of the gun, provided the mean gas-pressure be the same. If the projectile be of the same proportional length, and the gun .be also of the same proportionate length, and the muzzle velocity be the same, say 2000 feet in every case, then the time of travel up the bore will be directly in proportion to the length of the guns.
From which we deduce that the whole of the powder should be consumed by the time the projectile has completed .43 of its entire travel. Consequently the powder must be of such density and of such maximum thickness that it shall be consumed (under the pressure existing in the chamber and bore of the gun) at such a rate that the time of combustion shall equal the time taken by the projectile to complete .43 of its entire travel. But the time taken by the projectile to complete this distance will depend not only on the pressure of the gas, but also on the weight of the projectile and on the amount of the resistance of the rotating ring in passing into the rifled portion of the gun.
From an examination of the results of the experiments of Sebert and Hugonist and many more recent experiments, it would appear that the velocity of combustion of powder of very high density, when fired behind projectiles with normal rotating rings, must be much more rapid than that suggested by Piobert's assumed law and given by calculation in Table I. And from some other investigations which have been made (not given in the paper), it appears that the time of complete combustion of the charge of a 6-inch gun should not exceed 0.0062 second if the projectile weighs too pounds; and to obtain an absolutely uniform pressure during combustion, the weight of powder converted into gas should be proportional to the velocity of the projectile during the combustion of the whole charge. Therefore, if the velocity of combustion of the powder be constant under uniform pressure, the area of the combustion surface should increase in proportion to the increase of the velocity of the projectile during the combustion of the charge.
Now, it has been seen from Table I, in this paper, that the Quick cake powder is consumed in a much shorter time than the other powders, hence a very much higher density may be used and the whole charge properly consumed before the projectile completes more than .43 of its total travel. And this will give a low pressure, because the combustion and gas-generating surface is not at a maximum when the projectile is at rest and offering great resistance by its rotating ring, as it is in the case of P or prism powder. With a P powder of a high density and large size, the time of combustion would be so great that the grains would probably be blown out of the gun unburnt. See Noble and Abel's Researches, page 131. It is dangerous to use P powder of low density and large size, because it is found to produce abnormally high pressures, due probably to the large prisms being split up by the hot gases penetrating into them and converting them into small grains, and therefore producing extremely rapid combustion.
Again, it is equally dangerous to use P powder of high density and small size, because the initial ignition surface is so very large. For in a charge of say 100 pounds of powder composed of 2-inch cubes, the initial ignition surface would be roughly 4800 square inches, falling to zero as the combustion terminated, the reduction of surface being illustrated by Fig. 4, where, as also in Figs. 5 and 6, AO represents the area of initial ignition surface, OX the time of combustion, and the point X the area of the terminal combustion surface.
If the 100-pound charge be composed of 1-inch cubes, the initial ignition surface will be 9600 square inches, and the time of combustion at atmospheric pressure will be only 0.5 second, the reduction of the surface by combustion being illustrated by Fig. 5.
And if the 100-pound charge be composed of 1-inch cubes, the initial ignition surface will be increased to 19,200 square inches, and the time of combustion reduced under atmospheric pressure to 0.25 second. The process of combustion will be illustrated by Fig. 6.
With the cake powder there is complete control over the time of combustion of the whole mass (however high the density may be) by making the perforations more or less numerous, and thus varying the thickness of the powder burnt. Now the higher the density of the powder the more energy is stored up in a given volume, hence it is preferable to have the propelling material of high density so that it shall burn steadily and consequently generate the gas uniformly, the perforations being sufficiently numerous and close together to permit the whole mass to be converted into gas in the time required to obtain the maximum result in any gun of given length. The higher the density of the cake powder the less liability is there also for the cakes to be broken up into small fragments. Any such rupture of the cakes would of course affect the results. But as the whole of the cakes are ignited simultaneously, so every cake would be enveloped in its own atmosphere of fire, of uniform density and pressure, which would tend to protect every element or cake in the charge from being crushed up or ignited by impact with the other elements or cakes, for every cake would be equally and uniformly repelled by the others.
Hence we see the necessity for every gun of any given length to be supplied with powder suitable to its length and to the time the projectile is traveling up the bore.
Thus we may obtain a very high efficiency per pound of powder, and in consequence of the air-spacing being properly arranged, we obtain simultaneous ignition of the whole charge. Again, the initial ignition surface being reduced to a minimum, we obtain a low initial pressure, which enables us to reduce the total amount of airspacing in the chamber of the gun to a minimum. Then by reason of the charge being composed of a dense powder, with small airspacing, the charge will occupy but a comparatively small space in the gun, from which an economy will arise, as it permits of the gas pressure acting for a greater space and longer time on the base of the projectile. And most important of all, for naval service, is the fact that the smaller the size of the charge the greater is the number of charges that can be carried in the magazine. We thus approach the ideal but impossible propelling agent previously referred to.
There are some practical advantages in the use of the Quick cake powder when cemented, waterproofed, and made up into cartridges, which may be referred to.
It is well known that the hexagon or prism powder and the pebble powder cannot be made up into perfectly rigid cartridges for large guns, so as to fit the chamber with even a moderate amount of accuracy. The charges of the prism or pebble powder, instead of being cylindrical, are polygonal, having sharp angles, as shown in the annexed diagram, which is taken from the " Handbook of Artillery Material," by Captain Morgan, R. A. (Plate VIII, page 37).
And it can be seen from the Ordnance Reports, page 21, Part 1, ending March, 1887, that these "service" cartridges became unserviceable. It was reported from Gibraltar that on examining the charges, the Inspector of Warlike Stores "found the powder to be very dusty, and an amount of dust had worked through the bag." Some "prisms were much broken." "The cartridge was unserviceable." Suggestions have been made for the use of a stronger material for the bags. The authorities on board the Excellent, gunnery ship, considered that "as all heavy B. L. guns will be thoroughly washed out before reloading, strengthening the bags by stays of flannel or even leather would be admissible."
It is very evident that in the transport and handling of ordinary cartridges containing grain pebble or prism powders, a considerable amount of dust must be generated by the attrition of the powder, and .that the cartridge must become strained over the sharp edges or corners of the powder, so that the powder dust will work through the bags. It is thus easy to see that the smallest particle of fire in the gun will ignite this dust on the charge being inserted, the flash from which will ignite the whole charge.
Even in the case of perfectly new cartridges, it is highly probable that the cartridge bag is not unfrequently cut or torn on these sharp edges, on the charge being forced into the chamber, so that the powder may be uncovered in places, and thus rendered liable to be ignited by fire or friction against the hot metal or residuum in the chamber. It is probable that this is frequently the cause of the mysterious premature explosions which occur on reloading guns.
With the Quick cake powder the cartridges will be truly cylindrical, so that there can be no cutting or straining of the cartridge cases. The cakes being locked together by the clutch -like surface, there can be no attrition of the surfaces to generate dust, even if the cakes be not cemented together. At a very trifling expense, however, the whole of the cakes forming a charge or section of a charge may be cemented into one rigid mass, and the whole rendered waterproof and airtight. The front end of the cartridge may be protected by a shield or wad of non-inflammable material, so that any small amount of burning residuum would be swept forward by it out of the way of the powder, and thus premature explosion, even if fire remained in the gun, would be prevented. It is perhaps needless to say that the cylindrical form of the cartridge would greatly facilitate storage, handling and loading.
It may be reasonably believed that whatever development the nitro materials may have in the future, that their ballistic efficiency, keeping qualities, and their facility in use when made up into cartridges, will be much increased by being formed into cylindrical cakes with the central air-spacing, the whole charge being cemented together and made waterproof on the system herein described.