ply of sufficient albuminous matter is stopped. The relative proportions of these determine which of the two shall be first exhausted.
If the sugar is exhausted before the nitrogenous food of the fungus, a dry wine is produced; if the nitrogenous food is first consumed, the remaining unfermented sugar produces a sweet wine. If the sugar is greatly in excess, a vin de liqueur is the result, such as the Frontignac, Lunel, Rivesaltes, etc., made from the Muscat grape.
The varieties of grape are very numerous. Rusby, in his "Visit to the Vineyards of Spain and France," gives a list of five hundred and seventy varieties, and as far back as 1827 Cavalow enumerated more than fifteen hundred different wines in France alone.
From the above it will be understood that, cœteris paribius, the poorer the grape the drier the wine; or that a given variety of grape will yield a drier wine if grown where it ripens imperfectly, than if grown in a warmer climate. But the quantity of wine obtainable from a given acreage in the cooler climate is less than where the sun is more effective, and thus the naturally dry wines cost more to produce than the naturally sweet wines.
This has promoted a special cookery or artificial drying, the mysteries of which will be discussed in my next.
THE operations of benefit societies ought to provide a fruitful field for investigation by the student of vital statistics. These organizations, embracing a very large membership, deal practically with the accidents of sickness and of death, and can be made to supply valuable data. In Great Britain this has been done, and can be done, with comparative ease. The laws of that country take direct cognizance of benefit societies, and require from them statistical reports of their work, which are tabulated and published by authority. The latest available report of the British Registrar of friendly societies mentions no fewer than 15,379 branches of different organizations of this kind; of which number 12,300, embracing a membership of 4,672,175, had sent the required returns. The experience of so large a number of persons should certainly be of value. And we find that it has been utilized to a considerable extent, for from these returns tables of the expectation of sickness and death have been prepared by a number of English statisticians, including Ratcliffe, Finlaison, and others.
On this continent information concerning the operations of these societies can be obtained only from such reports as they publish themselves. They are not under Government supervision; and their sta-