Note On Attenuation

Note On Attenuation         Paper delivered 1898
by George Clissold

Until last April, I had been brewing for three years in a brewery where considerable difficulties were experienced for several years with the yeast. These difficulties had been attributed to the large quantity of nitrates in the water, which on several occasions was found to contain between 6 and 10 grains of nitric acid (as N₃O₃) per gallon. At the same time the ammonia was quite low, and the water very hard, containing about 120 grains per gallon of total anhydrous solids.

My predecessor met with some success by employing a very fast system of fermentation with large quantities of barm and high heats. He was, however, obliged to obtain a change of yeast about once a month.

For some time I carried this method on, but always disliked it, regarding high fermenting heats as dangerous, especially in summer, whilst I am of opinion that frequent changes of barm prevent that uniformity of produce which is so important, besides placing a brewer so much at the mercy of those from whom the changes are obtained.

At first I tried lower beats, and smaller quantities of barm, with very little effect. Believing that invert-sugar has a weakening effect on yeast, I tried glucose, and the result was most satisfactory as far as the yeast was concerned, but the beers would not fine, and possessed a peculiar, disagreeable flavour, so that 1 had to give this up, but tried a smaller quantity of invert. I again lowered the quantity of barm, and used some yeast food, and also tried dressing the yeast some time before skimming with flour and salt, together and separately. At last, finding the fermentations growing a little slower, I introduced rousing, and by means of yeast food, bucketing, low sugar percentage, and the use of moderate quantities of barm and a little salt dressing to the fermenting vessels, some barm was successfully kept healthy for three months. I then obtained a change, and worked the two yeasts side by side for two months, but then abandoned the new in favour of the old. This was about May, 1896. In July I abandoned the salt dressing as there seemed to be a salt flavour to the beers. This, however, did not disappear, and I was inclined later to attribute it to the fact that water was percolating into our well from a new artesian well which was being bored along side, the water in which was found to contain 65 grains per gallon of chlorides. However, that flavour gradually passed off, and disappeared when more sugar was used in the wort.

Meanwhile the general flavour of the beers had decidedly improved. The yeast was keeping quite healthy, and was working much slower; the beers were clean, and the attenuations were considerably lower. Thus, at the beginning of this change of yeast, in January, beers brewed at 22 to 23 lbs. (61—64°) attenuated as far as 5 to 4·4 lbs. (14—12°), that is, to about one-fifth of their original gravity, whereas beers of the same gravity were racking in November at 7·2, 7·7, and even 7·9 lbs. (from 20—22°), thus being well up to 1/3 O.G.

This very extreme difference might theoretically be due to different malt or to some difference in the mashing. That it was not so due, the following experiences will show.

In the autumn I had some slight difficulty in getting the beers to fine readily, aggravated, no doubt, by a sudden occurrence of cold weather. I never satisfactorily explained the cause of this difficulty, but in trying to overcome it I increased the sugar percentage again, and also tried a change of yeast, from the same brewery from which I had obtained that I had then been using for eight or nine months. The sugar made a considerable difference to the brightening of the beer, but the change of yeasts had no beneficial effect. This new yeast was worked side by side with the old for a month or more, and the difference between the attenuating power of the two yeasts will be seen from the results given in the table.

We may omit the consideration of the proportions of sugar used, as the results in (1) and (2) where the sugar is different are just the same as in (8) and (9) where it is the same, and the results of (6) and (7) are identical though (7) has half as much sugar again as (6).

It seemed to me that such great differences in attenuation and their results as regards condition, flavour, and fulness might be of interest to others as they were to me. In the matter of condition there was very little, if anything, to choose between the two yeasts. The beers were in fair condition, both quite bright, though both took from one to two days to fine perfectly, and both carried a nice head. As regards flavour, it is not surprising that there was a very considerable difference, hardly noticeable at racking, but developing in cask, so that after three weeks’ storage, it was impossible to doubt with which yeast the beer had been fermented. Unfortunately the flavour of the new yeast was decidedly inferior, and I was obliged to abandon its use. But it is remarkable that as regards fulness, as distinct from flavour, there was practically no difference.

These are all the details I wish to lay before you. The conclusions they seem to suggest are in the first place that attenuation depends far more on the species of yeast, and the conditions of its cultivation in as far as they tend to change the species—which must necessarily be a somewhat slow process—than on the malt or its manipulation in the mash tun.

With regard to flavour, the same may be said, and to an even greater extent, with the reservation that for certain classes of sweet ales a malt with plenty of fire is essential, whilst too much fire must not be allowed in a pale ale malt.

As regards fulness, the fact of two yeasts giving beers of such different specific gravity, but of almost identical fulness, seems to show that fulness is not dependent on the readily fermentable substances remaining in the beer, as it is hardly conceivable that the high attenuation should leave as much of such as is left by the low. On the other hand a beer that runs down very low generally tastes thin. I have myself experienced—and I have heard of others having the same experience—isolated cases of most unexpectedly and unaccountably high attenuation. These beers have tasted thin, and the yeast being the only factor in their brewing liable to any considerable variation, one is led to attribute both attenuation and thinness to its action.

Thus, whether it be dextrin, or whatever else which gives fulness, it is obvious that some yeasts can remove it, whilst others leave it. Though the pure yeast controversy is not yet defunct, I, for one, confidently hope to see the time when pure yeast will be in use in every brewery, and such difficulties as I have been recounting will be at an end.

I have been unable to give my reasons for many of the alterations I have described, as it would be tedious to go into them all. I hope, however, that in my endeavour to give so many facts, as briefly as possible, I have not made them unintelligible.