SELECTION FOR INCREASED BODY WEIGHT IN POULTRY. WHATROLE DOES SELECTION AGE HAVE ON RESPONSE?

NICHOLAS B. ANTHONY

CENTER OF EXCELLENCE FOR POULTRY SCIENCEDEPARTMENT OF POULTRY SCIENCE

UNIVERSITY OF ARKANSASFAYETFEVILLE ARKANSAS

ABSTRACT

Selection for rapid growth at a specific age has resulted in a tremendousgrowth response. Thus, in response to this success, the age at which selection isapplied has been adjusted to younger ages. The work described here is acomposite of several selection studies which were designed to obtain anunderstanding of the relationship between selection age and growth. The firstselection program discussed included the response to selection of 4 quail linesselected 8 generations for increased body weight at 10, 17, 28 and 40 days of age.Selection age had a profound effect on the magnitude of growth but had littleeffect on the sigmoid shape of the growth curve. In general, weight selected lineswere competitive with the other lines until the age at which selection was applied.After this time, early selected lines were smaller then the late selected lines. Twoadditional lines selected for rate of growth around the point of inflection (POI)were described. In general, selection for rapid growth prior to and slow growthafter POI (Line HL) effectively changed the pattern of growth in form withoutchanging asymptotic values. Selection for slow early and rapid late growth (LineLH) resulted in change in asymptotic values with little change in form. Line HLpossessed a compressed growth curve and was an early maturing. The variousselection strategies should provide insight into existing selection programs.

61

INTRODUCTION

The poultry industry evolved rapidly from the sale of live birds in themarket (1920s and 30s)to the further processed product of today. Althoughselection for rapid growth was being practiced, it was not until the early 1950'sthat the creation of line specific meat and egg type birds as well as events such asthe Chicken of Tomorrow Contest allowed the broiler growth potential to berealized (Gyles, 1989). Thus, within a relatively short period of time selection ageand market weights have shifted to earlier ages and heavier weights respectively.With this marked response has been the concomitant decrease in age at whichBW selection is applied (Figure 1).

It has been speculated that this shifting of selection age has led to some ofthe physiological complications being (ascites, tibial dyschondroplasia) observed intoday's rapidly growing poultry species. Katanbaf and coworkers (1988) reportedon the concept of resource allocation as it relates to weight selected lines ofchickens. Within this work, they described the timing of organ development as itrelates to growth. Work in our laboratory has found similar results for the turkey(Figure 2). In general, organs responsible for supplying nutrients have a higherpriority earlier in the growth period then those organs which have a relatively highnutrient demand. Thus, selection at younger ages may lead to emphasis of supplyorgans which would enhance the opportunity for lean mass deposition later. Thisconcept was supported by Lilja and Marks (1991) who reported that late stageembryos from weight selected of lines of Japanese quail had a higher proportionof gut tube then the randombred control from which the lines were initiated.

Growth curve analysis has been utilized as a means of understanding theeffects of selection for increased body weight on the pattern of growth of turkeys(Johnson and Gowe, 1962: Abplanalp et al., 1963; Anthony et al., 1991a), chickens(Tzeng and Becker, 1981; Zelenka et al, 1986) and Japanese quail (Marks, 1978;Anthony et al., 1986). Within these works is often a discussion of how selection

has altered the form (shape) and magnitude of the growth response (Asymptote).Ricklefs, (1967) provides an excellent description of growth curve analysis using 3parameter equations (Logistic, Gompertz and Von Bertalantfy). These equationsprovide information regarding form, rate and magnitude of growth throughprediction of points of interest on the curves, such as hatch, point of inflection(POI) and Asymptote. As a means of further understanding the effects ofselection for growth in various species a summary manuscript was prepared(Anthony et al., 1991b) which attempted to condense growth information fromturkeys, chickens and quail. One of the conclusions drawn from this work was

'The time of selection in relation to the POI may influence the response in BW".Unfortunately this work was confounded by factors such as generation number,selection intensity and environmental factors. However, it did provide the basisfor the selection trial to be described.

62

EFFECTS OF SELECTION AGE ON GROWTH RESPONSE

A selection experiment was initiated to study the effects of selection age onthe growth response observed. Japanese quail were utilized as the animal model.The selection ages chosen were based on the relationship of selection age andrespective age at POI for commercial/experimental poultry populations. Forexample, line H10 was selected for increased BW at 10 days of age andrepresented the relationship between selection age and POI for experimental linesof chickens (Anthony et al., 1991b). Line H17 represented selection for rapidgrowth at POI and the relationship between selection age and POI for commercialbroilers (Barbato, 1991). Line H28 was selected for high BW at 28 days andrepresents a relationship observed in turkeys (Anthony et al., 1991b). Line H40represents selection for increased body weight at mature weight. Initialheritabilities (parent offspring regression) for body weight at 10, 17 28 and 40days were .32, .29, .28 and .30 respectively. These heritabilities were consistentwith that observed for Japanese quail (Sefton and Siegel 1974) and other poultryspecies (Kinney, 1969) and indicated that response should be consistent betweenlines.

All lines were reproduced using a 36 pair mating system consistent with thesystem described by Nestor et al. (1982). In addition to the weight Selected lineswas a Random Bred Control (RBC) which was maintained with the weightselected lines. Approximately 400 progeny per line were reproduced eachgeneration.

During generations 6-8 the lines were surveyed for growth characteristics.In Generation 7, growth curve data were generated through alternate dayweighing of a random sample of quail. Data were exposed to analysis using theGompertz equation. Figure 3a,b provides the general growth curves observed forthe lines. Selection for rapid growth at any age was effective in increasing bodyweight when compared to the random bred control. It is interesting to note,however, that the response was not the same under each of the selectionstrategies. In fact, the selected lines were competitive with each other only to thatpoint where selection was applied. After the respective age of selection, growthwas found to be inferior to the lines selected at later ages.Realized heritabilities were calculated after the eighth generation of selection.The h2 were based on cumulative selection differential and selection responseexpressed as a deviation from the RBC (Figure 4). The pattern of h 2 estimateswere different from the original estimates predicted in the base population butwere consistent with the growth patterns observed. The relatively low h 2 observedfor the H10 line was likely due to factors such as scaling and residual maternaleffects.

Growth curve analysis revealed that the major adjustments in growth werenot in traits associated with form but rather magnitude (Table 1). For example,

63

all weight selected lines were lower than RBC for age at inflection and 90 % ofasymptotic weight. Body weight at these points, however, were only greater forH40 then the other lines. Slope at POI was not different for any of the weightselected lines but all greater then the RBC.

When weights are presented for each of the lines at each selection age(Table 2) one can see that all of the weight selected lines were heavier than theRBC at each of the sample ages. At 10 days of age weight selected lines did notdiffer in body weight, however, by 17 days of age line H10 was lower in weightthen the other weight selected lines. By 28 days of age line H10 was smaller thanH17 which was smaller than H28 and H40. At 40 days of age H40 was clearly thelargest followed by H28. Lines H17 and H10 did not differ in 40 day BW,however had lower weight than the other weight selected lines. Similardifferences were observed for yield data collected for the line. However, whenthe same data were expressed relative to BW the differences were eliminated forboth supply and demand organs (data not presented). This finding indicates thatthe selection at different ages has not led to the emphasis of specific organs butrather a consistent increase in organ size regardless of selection age.

From an industry standpoint this work has significant implications. Poultrybreeders (turkey and broiler) must understand the future markets for theirproducts. The temptation to select at younger ages consistent with processing agecould lead to an inferior product if the market were to shift to heavier weights.The converse situation would be in favor of earlier selection ages because onecould produce a bird which would be competition at the age it were to beprocessed and easier to manage as a breeder because of the lower mature weight.Unfortunately, this selection experiment did not provide a great deal of newinformation regarding the sensitivity of the growth curve to change in form.Historically, selection programs have been simply adjusting the magnitude ofgrowth to certain points on the growth curve (Figure 1) without substantiallyaltering the shape of the curve.

SELECTION FOR THE MODIFICATION OF GROWTH PATTERN

As previously mentioned, rapid rates of growth have been linked, justifiedor not, to the incidence of physiological breakdowns (ascites, tibialdischondroplasia, obesity) observed in todays poultry species. Methods to curbthe incidence of such problems include the slowing of early growth throughnutritional means (Plavnik and Hurwitz, 1985; Plavnik et al., 1986; Shlosberg etal., 1991; Acar et al., 1995). Feed restriction of breeder flocks is a necessary evilin order to obtain acceptable reproductive performance. Restriction programs aredesigned to alter the shape of the growth curve without a negative impact onperformance. Can the same result be achieved genetically or is the shape of thegrowth curve rigid?

64

Attempts have been made to shift the shape of the growth curve throughapplying selection pressure for high body weight early and low body weight late inthe growth period (Gyles and Thomas, 1963; Merritt, 1974; Ricard, 1975).Unfortunately, these studies reported minimal success thus suggesting that suchstrategies were of little utility. This may have, however, been due to the age atwhich selection was applied. In most cases body weight selection was practiced atonly two ages and both being post POI.

More recent attempts to modify the shape of the growth curve werereported. Barbato et al., (1992) described a divergent section program in chickenswhich utilized exponential growth rates to 14 and to 42 days posthatch. Marks(1995) reported on selection for body weight gain in Japanese quail between 2and 4 weeks of age on split and complete diets. In both of these programs, a shiftin philosophy occurred in that rates of growth are being considered rather then ameasure at a single point in time.

A second set of weight selected lines were developed as a means ofinvestigating the flexibility of the growth curve. The strategy was to place as muchpressure on the rates of growth around the POI as to develop two lines whichwould differ in form. To accomplish this two lines were developed. Line LH wasselected for low body weight gain between the ages 10 and 17 days and high bodyweight gain between 17 and 28 days of age. Line HL was selected for high growthbetween 10 and 17 days and low growth between 17 and 28 days of age.Preliminary work had indicated a correlation of -.13 between 10 - 17 and 17 - 28day body weight gains, therefore, selection of this type would at least becompatible.

Figure 5 includes a summary of male growth data for Lines HL, LH andthe RBC collected after 7 generations of selection. Line HL is characterized byrapid early growth and low late growth. Converse to this Line LH grew similar tothe RBC until 17 days and then began to accelerate. Growth curves of Lines LHand HL intersect at approximately 28 days of age. This point of intersection willprovide an interesting point of comparison between the two selected lines becausealthough the two lines are the same BW they achieved this weight differently.Growth curve analysis using the Gompertz equation revealed some interestingdifferences (Table 3). Line HL appeared to have deviated from the RBC forthose traits associated with form while Line LH was consistent with the RBC for

most of the traits measured except that associated with asymptotic weight. LineFIL had significantly more abdominal fat on an absolute and relative basis (Table4) and was at a more advanced stage of reproductive development (males, Table5; females, Table 6) at 40 days of age then either Line LH or the RBC fromwhich it was derived. In effect, the growth curve of Line HL has beencompressed thus resulting in a bird which grows rapid early and is early maturing.Despite a 16% difference in body weight between Line HL and RBC at POI therewas no difference between the two lines at maturity (Table 7).

65

It appears that there is some flexibility in the shape of the growth curve.However, shaping the growth curve is labor intensive in that multiple measures ofweight must be accumulated prior to selection. Unfortunately, the benefits ofgenetic manipulation of the growth curve are unclear at this time. Hopefully thework presented at this meeting and in future publications investigating selectionfor growth rates will provide the foundation for future selection schemes.

REFERENCES

Abplanalp, H., F.X. Ogasawara, and V.S. Asmundson, 1963. Influence ofselection for body weight at different ages on growth of turkeys. Br.Poult. Sei. 4:71-82.

Acar, N., F.G. Sizemore, G.R. Leach, R.F. Wideman, Jr., R.L. Owen, andG.F. Barbato, 1995. Growth of broiler chickens in response to feedrestriction regimens to reduce ascites. Poultry Sci. 74:833-843.

Anthony, N.B., D.A. Emmerson and K.E. Nestor, 1991a. Research Note:Influence of body weight selection on the growth curve of turkeys.Poultry Sei. 70:192-194.

Anthony, N.B., K.E. Nestor, and W.L. Bacon, 1986. Growth Curves of Japanesequail as modified by divergent selection for 4-week body weight. PoultrySci. 65:1825-1833.

Anthony, N.B., D.A. Emmerson, K.E. Nestor, and W.L. Bacon, 1991b.Comparison of growth curves of weight selected populations ofturkeys, quail, and chickens. Poultry Sci. 70:13-19.

Barbato, G.F., 1991. Genetic architecture of growth curve parameters inchickens. Theor. Appl. Genet. 83:24-32.

Barbato, G.F., 1992. Divergent selection for exponential growth rate atfourteen or forty-two days of age. 1. Early responses. Poultry Sci.71:1985-1993.

Gyles, N.R., and S.E. Thomas, 1963. Divergent selection for eight-weekbody weight in White Wyandottes. Poultry Sci. 42:1273-1274.

Gyles, N.R., 1989. World's poultry science invited lecture. Poultry Sci.68:1-8.

Johnson, A.S., and R.S. Gowe, 1962. Modification of the growth pattern ofthe domestic turkey by selection at two ages. Pages 57-62 in: Proc.XIIth World's Poultry Congress, Sydney, Australia.

66

Katanbaf, M.N., E.A. Dunnington, and P:B. Siegel, 1988. Allomorphicrelationships from hatching to 56 days in parental lines and F1crosses of chickens selected 27 generations for high or low bodyweight. Growth Dev. Aging 52(1):11-22.

Kinney, Jr., T.B., 1969. A summary of reported estimates of heritabilitiesand of genetic and phenotypic correlations for traits of chickens.USDA-Agricultural Research Service, Agricultural Handbook No.363. United States Government Printing Office, Washington, DC.

Lilja, C., and H.L. Marks, 1991. Changes in organ growth patternassociated with long-term selection for high growth rate in the quail.Growth Dev. Aging 55:217-224.

Marks, H.L., 1978. Growth curve changes associated with long-termselection for body weight in Japanese quail. Growth 42:129-140.

Marks, H.L., 1995. Divergent selection for growth in Japanese quail undersplit and complete nutritional environments. 8. Progress fromgenerations 18 through 30 following change of selection criterion.Poultry Sci. 74:237-246.

Merritt, E.S., 1974. Selection for growth rate in broilers with a minimumincrease in adult size. 1st World's Congr. Genet. Appl. LivestockProd., Vol. 3, Ministry of Agriculture, Madrid, 951.

Nestor, K.E., W.L. Bacon, and A.L. Lambio, 1982. Divergent selection forbody weight and yolk precursor in Coturnix coturnix japonica. 1.Selection response. Poultry Sci. 61:12-17.

Plavnik, I., and S. Hurwitz, 1985. The performance of broiler chicks andfollowing a severe feed restriction at an early age. Poultry Sci.64:348-355.

Plavnik, I., J.P. McMurtry, and R.W. Rosebrough, 1986. Effects of earlyfeed restriction in broilers. 1. Growth performance and carcasscomposition. Growth 50:68-76.

Ricard, F.H., 1975. A trial of selecting chickens on their growth curvepattern: experimental design and first generation results. Ann.Genet. Sel. Anita. 7, 427.

Ricklefs, R.E., 1967. A graphical method of fitting equations to growthcurves. Ecology 48:978-983.

67

Sefton, A.E., and P.B. Siegel, 1974. Body weight relationships of newlyhatched Japanese quail. Poultry Sci. 53:1254-1256.

Shlosberg, A., E. Berman, U. Bendheim, and I. Plavnik, 1991. Controlledearly feed restriction as a potential means of reducing the incidenceof ascites in broilers. Avian Dis. 35:681-684.

Tzeng, R.¥., and W.A. Becker, 1981. Growth patterns of body andabdominal fat weights in male broiler chickens. Poultry Sci.60:1101-1106.

Zelenka, D.J., E.A. Dunnington, and P.B. Siegel, 1986. Growth to sexualmaturity of dwarf and nondwarf White Rock chickens divergentlyselected for juvenile body weight. Theor. Appl. Genet. 73:61-65.

68

Table 1. Growth Curve Parameters for weight selected quailas described by the Gompertz equation (_-+ SEM)

1 Relative Inflection AsymptoteLine hatch

acle "weicjht slope Age at90% weight

H10 .051 + .002 15.4+.7 b 47.5+ 1.4b.075+.003 a 45.4± 8.8b 129.1 4- 16.7b

H17 .051 + .001 !14.04- .3b 47.54- 0.7b.0804. .002 a 41.34- 4.7 b 129.2+ 10.3b

.4b aH28 .048± .001 14.6+ 49.7± 0.8b.080± .002 41.9:1: 135.2±4.3b 10.8b

H40 .049+ .001 15.8±.5 b 54.6+ 1.0a.075+.002a45.1+ 6.2b 148.44. 13.8a

a b b a bRBC .052±.002 18.5+ .8 47.1± 1.3 .064±.002 53.3± 9.4 128.1 + 16.6

a,b Means within columns with different superscripts differ (P< .05)

1 H10 = quail selected for increased 10 - day BW;H17 = quail selected for increased 17 day BW;H28 = quail selected for increased 28 day BW;H40 = quail selected for increased 40 day BW;RBC =Randombred control

Table 2. Body Weights (_" + SE) of male Japanesequail after 7 generations of selection

Line 1 Day(g)10 D_y)17 Da(yg_8 Da(yg_0

H10 28.7+ 0.8b 53.5+ 1.2b 87.3+ 1.1c 107.8+ 1.3c

H17 32.1 .+.0.9a 59.0+ 1.2a 95.1 4. 1.3b 113.2:1:1.5b

ab a ab bH28 31.1 ± 1.0 59.2:1.1.5 98.2+ 1.5 116.6 ± 1.5

H40 30.8 4. 1.0ab 60.2 4. 1.5a 99.8 4. 1.6a 124.6 4. 1.6 a

RBC 22.4 i 0.8c 41.5 + 1.5c 72.5 ± 1.8d 95.2 + 1.3 da,b,c,d

Means within columns with different superscripts differ ( P < .05)

1 H10 = quail selected for increased 10 - day BW;H17 = quail selected for increased 17 - day BW;H28 = quail selected for increased 28 - day BW;H40 = quail selected for increased 40 - day BW;RBC =Randombred control

59

Table 3. Growth Curve patterns for weight selected quail asdescribed by the Gompertz Equation _'..+_SEM).

1 Relative Inflection AsymptoteLine ! hatchwt a£1e weicjht slope . Acjeat 90% weicjht

LH .046+.002 b 19.1+.8 a 50.5+1.1 a .065+.002 b 53.3+2.0 a 137.3+2.9 a

a b b a b bHL .051+.001 15.7+ .7 43.9+1.5 .074+.003 45.8+ 1.9 119.5+4.1

a a ab b a alRBC .052+.002 18.5+.8 47.1+1.3 .064+.002 53.3+2.0 128.1+3.5

a,b Means within columns with different superscripts differ (P < .05)

i LH = line selected for low BW gain between 10 and 17 days andhigh BW gain between 17 and 28 days

HL = line selected for hgh BW ga n between 10 and 17days andlow BW between 17 and 28 days;

RBC = Randombred control.

Table 4. Abdominal fat measures (7+ SEM)for quail after 7 generations of selection

1 Male FemaleLine Fat (g) % fat Fat (g) % fat

b b aLH 0.68 + .07 0.63 + .07 1.12 + .15 0.95 + .12 a

a aHL 1.15 + .10 1.15 + .09 1.22 + .17a 1.02 + .13 a

b b b bRBC 0.69 + .10 0.72 + .10 0.79 + .07 0.73 + .06

a,bMeans within columns with different superscripts differ ( P < .05)

1 LH = line selected for low BW gain between 10 and 17days andhigh BW gain between 17 and 28 days;

HL = line selected for high BW gain between 10 and 17 days andlow BW gain between 17 and 28 days;

RBC = Randombred control.

70

Table 5. Male Reproductive Traits (means + SEM)for quail after 7 generations of selection

Line 1 Testes Day 40 BW ADJ BW

b a aLH 1.58 + .58 103.9 ± 1.8 102.3 + 1.8

a b bHL 2.12+ .54 98.4+ 1.7 96.3+ 1.7

b bRBC 1.44 + .61 95.2 + 1.3 93.8 + 1.2 b

a,bMeans within columns with different superscripts differ (P < .05)

1 LH -- line selected for low BW gain between 10 and 17days andhigh BW gain between 17 and 28 days;

HL = line selected for high BW gain between 10 and 17 days andlow BW gain between 17and 28 days;

RBC = Randombred control.

Table 6. Female Reproductive Traits (means + SEM)for quail after 7 generations of selection-

1 Follicles Ovary Oviduct Day 40 ADJ BWLine (n)

LH 1.5±.4 b 1.54+ .44b 2.10.,+..49b 113.8+ 2.5 ab 109.9± 1.9 a

HL 3.7 ± .4 a 4.05 :f..46 a 5.35 ± .37 a a b117.1 :f. 2.4 105.2+ 1.9a

b b b bRBC 1.5 ±.3 b 1.33± .31 2.30± .40 108.4+ 1.7 103.7+ 1.2

a,bMeans within columns with different superscripts differ ( P < .05)

1 LH -- line selected for low BW gain between 10 and 17days andhigh BW gain between 17 and 28 days;

HL = line selected for high BW gain between 10 and 17 days andlow BW gain between 17and 28 days;

RBC = Randombred control.

71

Table 7. Body Weights (means + SE) of malequail after 7 generations of selection

Line' D_gy)10 Day(g)17 Da(yg_8 Da(yg_0b ab a

LH 22.4 ± .9 44.4 + 1.6 79.6 ± 2.2 103.9 + 1.8a

HL 26.0 4- .9 a 48.0 4- 1.6a 79.4 4- 1.3a 98.4 4-, .7 b

RBC 22.4+ .8 b 41.5+ 1.5 b 72.5+ 1.8b 95.2+ 1.3 b

a.bMeans within columns with different superscripts differ ( P < .05)

1 LH = line selected for low BW gain between 10 and 17 days andhigh BW gain between 17and 28 days;

HL = line selected for high BW gain between 10 and 17 days andlow BW gain between 17 and 28 days;

RBC = Randombred control.

?2

Figure 1. Predicted patterns of growthfor the chicken of 1925 and thebroiler of 1950, 1975 and lggo.

s°°°] _ ,9_sI _ ,9_o oooIIO__

I /m° 2000

iO01o 5 I o 15 20 25

Weeks of Age

Figure 2. Relative growth ratesfor organs in Turkeys

0.30"1 _ _ Bone

"Im---- %Organs

0.'251 _ %Legmuscle%Breast Muscle

0.20 _ %Fat

rr_ 0.15'

0.I0

I

0.05

O.OC0 1 2 3 4 5 6 7 8 9 I0 11 12 13 14 15 16 17 le 19 20

Age (wks)

73

Figure 3. Growth Curves of male

Japanese quail selected for high bodyweight at I0 (HIO), 17 (HI7),28 (H28), and 40 (H40) days of age.

Figure 3a. Curve from hatch to 40 days14o1130"1,2o-1 i__ HlOM _110-] _ .,7M,ool_ .28. ___'--_oi -.- _o. ___

_ 50N 40

302O1

0 2 4 6 8 10 12 1416 182022242628303234363840

Age (days)

Figure 3b. Curve from hatch to 18 days.

70-

60 _ HIOM

----l----HI7M _'_

30'

20"

lO"

0 ' I I I I I I I I I I I I I I I l I

0 2 4 6 8 10 12 14 16 18

Age (days)

74

Figure 4. Generation means (deviation

from RBC) plotted against cumulatedselection differientials for selected

lines of quail.

Realizedh 240

HIO .19HI7 .25 /

g 2o

o 20 40 60 80 IOO

Cumulated Selection Differential (g)

Figure .5. Growth Curves of Japanese quailselected for high early (10-17 day), lowlate (17-28 day) gains (HL), low early highlate gains (LH) and RBC.

II0"

I00" _ RI_C11

8oZ

70

60so

4030

20

I0

0 " I " l " l " l " l " l - l " l " m " l " l " l " l " l " l " l - m " m - i - i

0 2 4 6 8 I0 12 14 16 1820 22 2426 28 3032 3436 38 40

Age (days)

75

Question: J. Arthur

Do you have an idea of how much more final body weight you would have had incomparison to the LH and HL lines if you had an HH line?

Response: N. Anthony

It would have been nice to have both a HH and a LL line to complement LH and HL.Unfortunately, space did not allow for the additional lines. Despite this, I think a HHline would have had growth resembling line H28 (at least for asymptotic weight).Lines LH and HL currently mature at the same weight as H10 and RBC, respectively.

Question: M. Grossman

There is evidence in chickens, especially in females as they reach sexual maturity, thatthere are multiple points of inflection. Is this also true in quail?

Response: N. Anthony

This is something that we have observed in chickens but not in the quail model. Wehave attributed some of this to the timing of fat deposition which occurs relatively latein Japanese quail and is somewhat consistent with sexual maturity. Broilers have amuch greater propensity to deposit fat throughout the growth period.

Comment: B. Muir

Selection for HL or LH in the way you described is equivalent to selection formaximum difference. In that case, maximum overall gain may be zero. To ensure apositive overall gain while retarding gain of a segment of the curve, a restrictedselection index would be preferable.

Response: A. Anthony

Preliminary growth curve analysis revealed a fair amount of variation for pattern ofgrowth. How do you exploit these differences? The most ideal situation would be toweigh all birds every day and select the desirable pattern. I chose to identify LH andHL birds by the rate of growth around the point of inflection. The restricted indexselection may have guaranteed a response but I am not sure we would be answeringthe same question.

76

Question: M. Akbar

The conclusion was made that, in broilers the decrease in the age of selection did notchange the proportion of various organs (e.g. breast, legs, etc.) in the carcass. Whatdo you think about the applicability of this statement in the other meat species (e.g.ducks)?

Response: N. Anthony

The conclusion made was based on a short term selection trial on Japanese quailselected for increased body weight at different ages. After 8 generations of selectionwe observed absolute change in organ weight. No changes in relative organ weightwere observed. I would expect similar responses for other poultry species maintainedunder a similar selection scheme.

77