- ii0 - PART IV DR. CARL W. HESS. U. S. DEPARTmeNT OF AGRICULTURE SUNDAY MORNING SESSION MAY 12, 1963

Transcription

1 - ii0 - PART IV DR. CARL W. HESS U. S. DEPARTmeNT OF AGRICULTURE SUNDAY MORNING SESSION MAY 12, 1963 DR. HESS: I believe we are ready to call the meeting to order this morning. I am glad to see so many of you here on Sunday morning. We have a rather close schedule this morning and we are about ten minutes late in getting started. As you know, we have two very excellent papers this morning, and so I will take very little time so as to allow as much time as possible to the speakers and for questions following the presentation. The first speaker this morning needs no special introduction, because Dr. Lush did a very fine job of presenting him yesterday morning, and all of us heard him give his presentation on Concealed Genotypic Variation. So without taking any further time from Dr. Eendel, I will call at this time on Dr. Rendel. His topic is GENETIC CORRELATION BETWEEN _,I0CHARACTERS IN DROSOPHILA. Dr. Rendel. 3

2 -lll- G_ETIC CORRELATION BETWEEN I_40CHARACTERS IN DROSOPHILA Dr. John M. Rendel Director of the Division of Animal Genetics of the Commonwealth Scientific and Industrial Research Organization of Australia In 1934 %fright put forward a model which was intended to account for dominance on physiological grounds. Suppose a substrate T is converted by an enzyme to product y with efficiency a and by another enzyme to some irrelevant product with efficience d_.we can then write: T a and the relative rates of the reactions, or the fraction of T converted in unit time is _ to Y and d to the irrelevant product. Since a appears in both a+d numerator and denominator, increase in the function of T converted to Y as a increases follows the law of diminishing returns, thus 2a is less than twice 2a+d a and if a is large compared to d, a soon reaches a value such that further a+d increase in a has little effect on the fraction. y If we suppose that the reaction going at rate d is not irrelevant (let us call it D) and that T is not constant, this model is one which could describe genetic correlation. As T increases the amount of Y and D increases. The correlation is positive. As a increases relative to d, d decreases sad the correlation is negative since Y increases and D decreases. The correlation illustrated here is of a kind in which two distinct physiological processes share a common resource. This model clearly has nothing to do with of the same correlations due to linkage or to the two characters being two measures single physiological processes. / ") I shall bring forward evidence to show that the model is in fact the right one for describing correlations due to the sharing of resources by more than one

3 - ll2 - physiological process. The two correlated characters are scutellar bristles and abdominal chaetae. The sex linked gene scute reduces both. It removes all or most of the scutellars and about half the abdominals_ On9 might expect that selection to counteract the effect of scute would increase both scutellar and abdominal bristle number. In the selection experiment already described, the abdominal bristles were counted after about 25 generations of selection on scute and the figure 1 shows the result. In sc males the abdominal bristle number has been increased but in sc/sc females there has been no change and in wild type sibs the abdominal number has actually fallen in the High selection line. Fig. 2 shows the effect of a substitution of sc for + and sc/sc for +/+ in the high and low lines. The slope relating scutellar to abdominal bristle number is much flatter in the high than in the low line. This can be interpreted as showing that a bigger share of T is going to scutellars and a smaller share to ahdominals in the high line. The slope which shows how much abdominals change for a given change of scutellars could be taken as a direct measure of the way T in our model is being shared out if the abdominals and scutellars were measured in equivalent units. Unfortunately we can say nothing about the equivalence of abdominal and scutellar bristles and this must first be determined. The method used to estimate the equivalence of scutellars and abdominals was as follows. The amount added by selection to T must be the same for all genotypes, if we subtract sc of one line from sc of the other, + from +, sc/sc from sc/sc, sc/+ from sc/+ and +/+ from +/+, all five answers should be identical in terms of T. But as the difference is composed of scutellars and abdominals, and the relative change in abdominals is positive in sc but negative in ++ we _-_ should get a regression of scutellars on abdominals the slope of which is the equivalence factor we are looking for. As both high and low lines were scored

4 - i13 - twice after different intervals of selection we have in all 6 comparisons which when reduced to a common mean are shown in Fig scutellar bristle is equal to 2_7 abdominals. One further step must be taken before the whole series of events can be clearly stated. We must find the value of scutellar score at which T is zero. Table 1 shows the scores of the _ counts. If the scutellar scores are reduced to the point where the abdominal score is zero and we take this as the point where T is zero_ we find we must add 1.33 to males and 3.05 to females. We can now convert table 1 to table 2o We find the best estimate of a, the function of T going to scutellars is.68 in the l_ lines and.56 in the high lines; (1 - a) is thus.32 and.hh respectively. This change in a accounts for some correlations being positive and some negative despite increases in T. Algebraically this can be expressed as follows. T is total substrate, a the fraction going to scutellars, N the addition to T due to selectioh, C the effect of substituting sc for +, a_ the changed value of the fraction going to scutellars after selection. Initial Then we have the following values for different male genotypes: After selection scutellar abdominal scutellar abdominal sc T(a) T (l-a) (T+M)_ (T+M) (l-al) + (T+C)a (T+C)(I-a) (T+C+M)a I (T.C+M)(l-aI) Since ai is greater than a and M is positive (T+M)al> Ta and (T+C+M) --T (T+C)a what we have to show is the conditions in which T(l-a) <T+M(l-al) and (T+C+)(l-a) / These are when T T+M _I-% l-a / and T+C /1-s. Since T+C is_ T, C being added to denominator and numberator it is / possible for T+C to be greater than _ when T is less than i i.e. for T+C+M l-a T+M l-a

5 > - 11h- + to show a negative when sc shows a positive correlation of scutellars and abdominals. A numerical example may help. Let T=2 M=4 C=18 a2_ ai--_then: Before selection After selection scutellars abdominals scutellars abdominals sc -,-,_2 = 1 _-'x2 = 1 _x (2+4) = 4 _(2+h) =.2 2 positive correlation + _x20 = IO _x20 = IO 2(20+4)=16 1(20+b)=8 negative correlation Practical conclusion from this work is that a 3 genetic correlation has sign and magnitude depending on whether selection is directed at T or a and on the level of expression of the characters correlated. The breeder may choose what sort of correlation he will have by using appropriate selection.

6 - ll5 - TABLE 1 Low line Low line High line High line scute- abdom- scute- abdom- scute- abdom- scute- abdomllars inals llars inals liars inals llars inals sc _ sc sc sc , _ Ii O TABLE 2 Low 2 Low I High i High 2 sc , io.66 sc sc sc _ o h a=o.54 a=o.68

7 _) - ll5a - O scsc99 abdominal...._ bristles oi/ sc _ I L I H FIG. 1 n_ abdominals _/..._/1_sc / + / _ High s_ sc0 Line Scutellars FIG. 2

8 _e _ "%e zx "_ Scutellars FIG. 3

9 DR. HESS: Before I ask for questions, I want to again remind you that those of you that ask questions should come to the center microphone and state your name and also speak up so that the reporter can get your name and your question. Does anyone have any question? I believe you have done such a very good _ob of presenting this that there are no questions. DR. RENDEL: DR. HESS: I tried to. If there are no questions, has anyone any comments? If not, then I am supposed to mention here that immediately after the break Dr. Briles wants to discuss the problem of _ether or not this meeting should be held at a different time next year, and also whether or not you want to hold this mee$ing on a week end, so you might think about it and maybe discuss it with some of your co-workers as to what your opinion is. I think maybe he wants to get a show of hands later on as to the number of people that could not be here maybe if this were not held on a week end. DR. BRILES: Since we have a little time before the coffee is actually ready, it was due at about the time of our break, we might just go ahead and take a show of hands here now. Has Don Turnbull arrived yet? He had a Sunday School class to teach before he came and was going to make the second session. Dr. McClary mentioned this to me yesterday. You have been in on these discussions. Would you care to come up and talk to the group and get their reaction and a show of hands with had in mind? regard to the type of policy matters that you DR. McCLARY: I suppose this is more in the direction of the eight-hour day and five-day week, or maybe, of course, less than that. But some have raised the question of whether we need now to reconsider this matter of week end meetings.

10 - i17 - I know my family likes to have me home on the week end; at least they tell me they do, and they also objected a little bit to this being Motherls Day away from home too. I suppose the first thing we should consider here is whether or not we feel it of sufficient importance to hold this immediately before the American Feed Manufacturers Association meeting. Now, as long as we are holding this meeting here in Kansas City and the others in Chica o, I think it is fixed, isn't it? Someone wondered if it was sufficiently important to coordina_e the two, since they are quite a distance apart. I think at this time it would be appropriate to a_ for a show of hands of those who wish the meeting to be held on the week end immediately preceding or at least to be held immediately preceding the American Feed Manufacturers Association. That would be those of you this would make a real difference to and would benefit you in the way of traveling, if you would prefer to have it on the week end before or only a few days before the American Feed Manufacturers Associatioh meeting. Could we have a show of hands then of those who have a strong feeling in this regard? DR. BELLES: DR. McCLARY: There are two hands. It seems then that this doesn't actus/ly involve a large number of people. We have in previous years, I think, held this closer to the last week end in April or the very first few days of May. I think that is the latest we have held it, with the exception maybe of o_ meeting. Then the other point, which is perhaps a more important one, we have considered holding Poultry Breeders of America meeting on Thursday. This is the meeting _-J of the Poultry Breeders themselves, a meeting to which we invlte and encourage

11 research and university personnel that are interested in poultry breeding procedures, business problems, and to some extent other factors surrounding the poultry breeding operations as they would affect all breeders. That then would leave Friday and Saturday open for the meeting that we have normally started on Saturday morning and ended Sunday noon. Now, we don't want to do anything here that would discourage the university people from participating in this meeting because they certainly are a valuable part of it. We couldn,t hold a meeting without them, in fact. So your wishes will certainly weigh heavily in determining whether this change is made or not. Now I suppose the procedure here would be to have a showing of hands of those people that could not participate if we held the National Breeders Roundtable on Friday and Saturday, that could not participate or it would cause a hardship if they did. Let's take the National Breeders Roundtable first. Let's see the hands of those that would like to continue participation and that could possibly not be here sometime if we held it Friday and Saturday. _ay we have your hands, please DR. BRILES: Three hands. DR. McCLAEf: Now I guess the next thing is the Poultry Breeders of America meeting on Thursday. If we hold that on Thursday, I notice there are a number of college people here who do attend this meeting with us. Those that normally attend and would like to attend this meeting, how many could not attend if we held this meeting on Thursday? May we see your hands, please. Thank you. I believe that gives us the information that the Board will need when they consider this matter. - DR. HESS: I think it may be just a few minutes before the coffee is ready,

12 but we are supposed to meet back here at ii o,clock, so we will have our coffee break at this time, and again I want to thank Dr. Rendel for presenting this paper to us. Thank you very much. (Coffee break) DR. HESS: Our next speaker also was introduced yesterday, so I don't think it is necessary to go into much detai_. However, I might mention that I visited with him a few minutes ago and I think some of you might be interested in the fact that he obtained his degree at the University of Texas, and that later on he spent about eight years at the University of Michigan before he returned to the University of Texas. Yesterday he talked on qualitative gene action, and today he is going to talk on quantitative gene action. So I will now call on Dr. Sutton.