lbs. Gross 133 Tare (deducted) 37 ----- Actual Tea paid for 96 pounds, on which duty is also paid. ===== Therefore the loss on chest is 2 ounces only. Now did weights turn out the same in London that they were on the garden, we could, by doing as in last example, insure only the above trifling 2 ounce loss. But it is _not_ so. The wood dries and thus makes both the gross and tare less. The loss then comes out anything between 2 ounces and 1 pound 14 ounces. I find the following simple rule will give the exact loss on each and every weight of both gross and tare. _Rule._--Add the ounces above a pound in the gross to the ounces short of a pound in the tare. The sum of the two, in ounces, will be the loss of Tea on the package. This is only part of the article. I break off here to add a few remarks more appropriate now than what I then wrote. There are means by which this varying loss, of which the maximum is 1 pound 14 ounces, can be reduced to 4 ounces only on each and every chest. I admit the procedure is scarcely practical, but as nothing can demonstrate better the absurdity of the system as pursued at the Customs, I give it here. How can we insure the _least_ loss, taking into consideration the fact that the weights of both the gross and tare, because of the wood drying and lightening in transit, can never come out the same at the Custom House in London as they were on the garden. We can do it thus: the Tea if well packed in a chest in no way alters in weight during transit. If dry, when put up, it cannot become lighter; if the leaden covering is air-tight, it can absorb no moisture, which would of course make it heavier. I therefore beg the question that it is a _fixed_ quantity, for it must be so if well packed. We have therefore only to consider the gross and the tare, and, as shown, the loss in Tea, varying from 2 ounces to 1 pound 14 ounces, depends entirely on the weights these are found to be at the Custom House. In other words, if we can insure the gross there being but little over any even number of pounds, and the tare there being but little below any other even number of pounds, we attain (approximately) the least loss we can be mulcted in. Begging the question that we can add to, or detract from, the gross weight of each chest in the Custom House (before it is put into the scales by the officer there) by the addition or subtraction of a few nails if the weight is nearly what we want, or pieces of hoop iron if the actual varies much from the desired weight--I say, if we can do this, we can insure approximately the minimum of loss. I go to show how this is to be done.