There is nearly always some loss of coal due to moving it.  I say ‘nearly always’ because it seems that there are occasions on which coal being moved increases in bulk.  It occurs when competitive coaling is being carried on in a fleet and ships try to beat records.  A collier in these circumstances gives out more coal than she took in.  We shall probably be right if we regard the increase in this case as what the German philosophers call ‘subjective,’ that is, rather existent in the mind than in the external region of objective, palpable fact.  It may be taken as hardly disputable that there will be less loss the shorter the distance and the fewer the times the coal is moved.  Without counting it we see that the annual expenses enumerated are—­

Establishment charges L6,500
Landing and re-shipping 5,060
Deterioration 3,795
-------
L15,355

This L15,355 is to be compared with the cost of the direct supply system.  The quantity of coal required would, as said above, have to be carried in twenty colliers—­counting each trip as that of a separate vessel—­with, on the average, 2300 tons apiece, and five smaller ones.  It would take fully four days to unload 2300 tons at the secondary base, and even more if the labour supply was uncertain or the labourers not well practised.  Demurrage for a vessel carrying the cargo mentioned, judging from actual experience, would be about L32 a day; and probably about L16 a day for the smaller vessels.  If we admit an average delay, per collier, of eighteen days, that is, fourteen days more than the time necessary for removing the cargo into store, so as to allow for colliers arriving when the ships to be coaled are absent, we should get—­

20 X 14 X 32 L8,960
5 X 14 X 16 1,120
-------
L10,080

as the cost of transferring the coal from the holds to the men-of-war’s bunkers on the direct supply system.  An average of eighteen days is probably much too long to allow for each collier’s stay till cleared:  because, on some occasions, ships requiring coal may be counted on as sure to be present.  Even as it is, the L10,080 is a smaller sum than the L11,560 which the secondary base system costs over and above the amount due to increased deterioration of coal.  If a comparison were instituted as regards other kinds of stores, the particular figures might be different, but the general result would be the same.

The first thing that we have got to do is to rid our minds of the belief that because we see a supply-carrier lying at anchor for some days without being cleared, more money is being spent than is spent on the maintenance of a shore depot.  There may be circumstances in which a secondary base is a necessity, but they must be rare and exceptional.  We saw that the establishment of one does not help us in the matter of defending our communications.  We now see that, so far from being more economical than the alternative method, the secondary base method is more costly.  It might have been demonstrated that it is really much more costly than the figures given make it out to be, because ships obliged to go to a base must expend coal in doing so, and coal costs money.  It is not surprising that consideration of the secondary base system should evoke a recollection of the expression applied by Dryden to the militia of his day: