Potential usefulness of the Na (Naked Neck) gene in poultry production

The possible interpretation of these effects is discussed for each trait. Concerning the potential usefulness of the Naked Neck gene in poultry production, present-day results show on the whole an encouraging perspective at ambient high temperature, either for broiler or for egg production. At low or moderate temperatures an obvious defect associated with the Naked Neck gene is a deterioration of feed efficiency, either for growth or for laying, but the results cited suggest that this defect may be suppressed by a more precise control of the environmental temperature and possibly, as concerns egg production, by combining the Naked Neck and the sex-linked dwarf genes.

(1) from day-old toslaughter age; (2) Not given. "Starter" ration to 4 wks, then "finishing"; (3) with fluctuations; (4) constant (i- no statistical test (measurement as a roup for each genotype); (6) not tested within environment, but G x E interaction signiticant for body wt (***) and feed efficiency (**? Only the data of Hanzl and Somes (1983) showed a marked inferiority of Nuked Neck homozygous and heterozygous females and of heterozygous males as compared to the na+ na+ genotype. Conversely, in the two cases where it was measured at moderate temperature, the feed efficiency of Nbked Neck chickens deteriorated in comparison with normally-feathered chicks, as might be expected from the larger extension of unfeathered areas, increasing the heat loss of the former.
At 24°C (Zein-el-Dein et ul., 1984) or at 25°C k 1°C (Merat, unpublished) differences in growth rate were small and insignificant, and the disadvantage shown for Nuked Neck bird feed efficiency at lower temperatures did not appear. Finally, when approaching or exceeding 30"c', the performances of Nuked Neck homozygous or heterozygous chickens are consistently superior to those of nu+ nu+ birds of the same origin. At 29°C (Zein-el-Dein et al., 1984a) this advantage was limited and non-significant for body weight; feed efficiency seemed slightly better although not significantly, for the Nuked Neck genotypes; in addition, the data suggested that the advantage is more marked when a ration with a lower protein level is given. At 31"C, the superiority for growth rate to 10 weeks of age, observed by Bordas et ul. (1978) on males and by Monnet er al. (1979) in both sexes for both Nu Nu and Nu n u f genotypes, was significant and attained more than 10 per cent. In the experiment of Hanzl and Somes (1983) at 38"C, Nuked Neck homozygotes showed a marked advantage for body weight which was less for heterozygotes. Feed efficiency was slightly, but not significantly, better for Naked Neck chicks at 31°C and 38"C, respectively, (Bordas et ul., 1978;Monnet et ul., 1979;Hanzl and Somes, 1983). Table 3 summarizes published data on body composition and yield of Nuked Neck chickens compared with chicks with normal plumage. The table includes the three criteria showing the most conspicuous differences between the genotypes: feathers as per cent of body weight, eviscerated carcass as per cent of live weight and meat yield as per cent of eviscerated carcase weight. As in Table 2 the deviations of the Nu Nu and Nu nu+ genotypes from the mean value of the nu+ nu' genotype are given, but here this deviation represents a difference on a variable already expressed in percent.

Body composition, slaughter yield and meat yield
For these traits the difference between genotypes do not seem to depend on environment conditions such as temperature. Except in one case, the proportion of feathers of Nu nu', and even more of Nu Nu, birds was consistently and highly significantly lower than that of nut nu+ chickens. The reduction of the plumage in proportion to live weight was of the order 1.5 to 2% and 2.5 to 3% for the Nu nu' and Nu Nu genotypes respectively. This may explain entirely the improvement of slaughter yield associated with these genotypes, which is regularly found although it is not always significant.
Another consistent highly significant difference in the data of Zein-el- Dein et ul. (1984b), concerns meat yield (percentage of muscles) of the eviscerated carcase which is superior for the Nu nu' genotype compared with n u f nuc (Nu Nu was not included in this experiment). El-Attar and Merat (1985) indicated that this difference is located in the trunk and probably is due to breast muscles. The advantage of the Nu na+ chicken for meat yield is variable. The results from Table 3 suggest that it may be relatively higher in medium or light-weight populations, and possibly also in those given a low protein diet (Zein-el-Dein et al., 1984b). This table does not include traits associated with fatness. For abdominal fat as defined according to     (2) Not gtven. "\ti~rter" ration 10 Ricard and Rouvier (1967) and Bordas et al. (1978), on male chickens raised at 31"C, did not find significant differences associated with genotypes at the Nu locus; a tendency to a slightly higher value for Nuked Neck birds might reflect their higher growth rate at this temperature. On the other hand, Zein-el-Dein ef al. (1984 b) and El-Attar and Merat (1985) measured abdominal fat and subcutaneous and intermuscular fat in the eviscerated carcase. Table 4 shows their results in a comparison of Nu nu+ and nuf nuf birds for these criteria. x ,2070 protein in feed

56
While abdominal fat does not differ according to genotype, the total of the percentage of subcutaneous and intermuscular fat is significantly lower in Naked Neck heterozygotes in the data of Zein-el-Dein et al. (1984 b), and the trend is the same, although not significant, in results for El-Attar and Merat (1985). This may be related to the analysis of total lipids in the carcase made by Hanzl and Somes (1983) who showed a significantly lower value for the Nuked Neck genotypes than for wholly-feathered birds.
Finally, Zein-el-Dein et al. (1981) and Zein-el-Dein et a f . (1984b) found that the head as a percent of live weight was significantly reduced by the Na gene; Zeinel-Dein et af. (1984b) in females and El-Attar and Merat (1985) in both sexes found a reduction for the neck in proportion to live weight and Zein-el-Dein et a f . (1981) drew a similar conclusion for shanks in females. These parts contain a high proportion of bone and may be compared to the higher proportion of muscles and lower proportion of skeleton found consistently in the eviscerated carcase of Naked Neck birds (Table 3).

Effects on egg production traits
The only available data on the effects associated with the Nu gene on egg production traits are those obtained by Bordas et al. (1980) and Monnet et ul. (1980) in a normal sized (Dw) population, and those presented by Merat and Bordas (1974) and Bordas and Merat (1984) evaluating the combined effects of genotypes at the Nu and D w (sex-linked dwarfism) loci. The recent unpublished data of the latter authors are included. Horst (1980) also reported briefly that at high ambient temperature the Naked Neck gene is associated with a 7.4% gain in total egg mass during the first 3 months of production, while Smith and Lee (1977) mention the lack of significant differences for egg laying and mean egg weight between Nu nu+ and nu+ nu+ genotypes at moderate temperature. Table 5 condenses the published data comparing the 3 genotypes at the Na locus for parameters related to egg production and feed efficiency. Both cited works include a "control" and a "heated" group. The results of both experiments are similar for most criteria. In the control group, an appreciable depression of adult body weight is associated with the Na Nu genotype in both normal-sized (Dw) and sex-linked dwarf (dw) hens (around 15%) and to a much lesser extent with the Nu nu+ genotype'. In spite of this, average egg weight is slightly increased for Naked Neck hens (significantly in the first experiment on a D w strain). The same appears for total egg mass produced per 28 days. In contrast, egg number and percent of cracked eggs do not differ significantly according to genotype at the Nu locus, although according to other unpublished data the Nu Nu hens might produce slightly more shelless eggs. In the group at higher temperature, there were no differences between genotypes for body weight. Conversely, the superiority of the Nu nu' and to a larger extent of the Nu Na genotype for mean egg weight and egg mass is more marked than in the control and is always highly significant for egg weight which was higher by more than 3 grams in Nu Nu hens than in the nu+ nu+ genotype. As in the control group, egg number and the percent of cracked eggs did not show any appreciable difference between genotypes, although for this last trait an advantage of Nu Nu layers at high temperature is suggested.
Feed efficiency is the only criterion for which the genotypes at the Nu locus do not differ in the same way in a normal-sized ( D w ) strain (experiment 1) and a dwarf ( d w ) strain (experiment 2). In the first case, although the differences in feed efficiency are not significant, the Nu Nu hens were the least efficient in the control environment, easily explained by their excessive heat dissipation. This handicap is not found in the high-temperature group. Among dwarf hens the feed efficiency was always better (or at least equal) for the Nuked Neck genotypes compared to the nu+ nu+ genotype, whatever the environmental temperature. Bordas and Merat, (1984) suggested that the reason for this is different depending on the ambient temperature. In the "control" at moderate temperature, it was due to the reduction of Nu Nu and Nu nu+ body weight without any depression in egg weight, but at a higher temperature it was due to the larger increase in mean egg weight and the egg mass for these genotypes without a proportional increase in feed intake. On the other hand, the deviation from a multiple regression equation, called "residual" food consumption at fixed body weight and egg production (Bordas and Merat, 1984), did not differ at high temperature. On the contrary, the differences between the genotypes Nu Nu, Nu ylu+ and nu+ nu+ in the control environment were much larger in the experiments with D w females than with dwarf hens (Bordas andMerat, 1980, 1984).
I n order to evaluate the effects associated with the Nu Nu and Nu nut genotypes at a temperature intermediate between those realized in the' previous experiments, a new comparison was undertaken at a constant temperature of 25°C from the age of 18 weeks, the mean daily temperature between 4 and 18 weeks varying approximately between 15 and 20°C. A dwarf ( d w ) strain segregating for the Nu gene was used. Table 6 shows the main results (unpublished). In this environment, the effect of the Nu gene on adult body weight was less than at lower temperatures, the differences between Nu Nu and izu+ nu' females decreasing between 18 and 39 weeks of age. The difference in mean egg weight showed the same trend but less than at 30°C. There were no appreciable differences between genotypes for feed efficiency so that at 25°C the Nu Nu and Nu nut genotypes as compared with nuf nuf had only a 1 g advantage in mean egg weight.
Finally, the possibility that the increased feed consumption associated with the Nu gene might have a beneficial effect in the presence of a ration with a suboptimal total protein level was investigated. This unpublished trial used a dwarf population with the Nu allele segregating and with 12% total protein in the feed from the age of 18 weeks. The ambient temperature was 25°C from the same age. The results are shown in Table 7. No differences between genotypes at the Nu locus appeared for body weight at 39 weeks and the differences in average egg weight were insignificant, but the trend was of the same order as in the previous experiment ( Table 6 ) . Laying rate and feed efficiency for egg production did not differ significantly according to genotype, but the egg mass produced in 28 days showed an advantage for Nuked Neck layers, significant at the 5% level. This effect and the effect on mean egg weight however was not higher than with the higher protein level given in the experiment corresponding to Table 6. On the whole, a significantly higher value of the Nu Nu genotype over the normal homozygote for semen volume and number of spermatozoa at various ages appears, the Na na+ genotype being intermediate. There is no significant interaction between genotype and temperature. On the other hand, the differences at the level (1) Hens inseminated by mixed semen from of semen concentration are not reflected in fertility. This trait, however, was estimated only from the pooled semen of 5 males per genotype and environment, and no statistical test was possible. Crawford (1977Crawford ( , 1978 published data on hatching percentage according to the genotype of the embryos at the Nu locus. He concluded a slight disadvantage for the Nu Nu and to a lesser extent, the Nu nu' embryos. Horst (1980)  These postulated effects on hatching rate may not suffice to explain certain deviations to Mendelian proportions, previously observed and limited to one sex (Merat, 1970), for which a hypothesis of selective fertilization was suggested.

Post embryonic mortality
There are no published data on the mortality of Nuked Neck compared to normallyfeathered birds at different ages. Horst (1980) mentions (no figures given) that the Nuked Neck gene seems to have a positive effect on chick viability at high ambient temperature. In the experimental conditions that achieved during the growth period (Bordas et ul., 1978;Monnet et ul., 1979) or at the laying stage (Bordas et ul., 1980;Monnet et ul., 1980), mortality was relatively low and did not differ according to the genotype at the Nu locus.
Conversely, in the presence of a high sublethal temperature (heat stress), Smith and Lee (1977) observed a significantly higher survival rate for Nu nu+ chicks (51.4%) than for the "normal" nu' nut genotype (38.8%) In an unpublished result a similar difference was obtained between male chicks of these two genotypes submitted to a temperature exceeding 40°C for several hours at the age of 2 days (heterozygous Nuked Neck, 59.3% survivors among 108 hatched: normally-feathered chicks, 33.3% survivors among 105 hatched; P<0.001).

Other effects
A consistent morphological effect noted in Nuked Neck birds (Monnet et ul., 1979;Monnet et ul., 1980;Bordas et ul., 1984) is a significant 10 to 20 percent increase of wattle size in both sexes and at all ages. This effect is local, as comb size is not higher for Nuked Neck cocks compared to their fully-feathered counterparts (Hammade et ul. 1986a).
The cloaca1 temperature of Nuked Neck birds is slightly inferior on average to that of nu+ nu' adult males and females; irrespective of the ambient temperature, the difference is about 0.1 to 0.2"C (Monnet et ul., 1979;Monnet et ul., 1980;Bordas andMerat, 1984. Hammade et ul. (1986a) make the same observation on cockerels or cocks at different ages. Monnet et ul. (1980) find a difference of the same type for the surface temperature of the shank, this difference being higher at 30°C than around 20°C. Hammade et ul. (1986a), notice a similar effect on skin temperature in males at two ambient temperatures, 18°C and 30°C.
According to Monnet et ul. (1980) in laying hens and Hammade et ul. (1986a) in males at several ages, the packed cell volume of Nu Nu and Nu nu+ birds is higher than that of nut nu+ individuals at 30"C, but not at 18°C. Monnet et ul. (1980) mention that the water intake of Nuked Neck hens is lower than that of fully-feathered hens at a high temperature; this may be attributed to their ability to dissipate heat into the environment more rapidly.
Zein-el-Dein et ul. (1984~) observed a difference in the behaviour in several respects between Nuked Neck and normally-feathered chickens in individual cages between the ages 1 and 10 weeks. Nuked Neck chickens spent the same time at the troughs as their fully-feathered sibs but they stood up more often. Conversely, in 2 day old chicks in the open field, the Nu Nu genotype had a longer latency time before the first move and moved less in a given time than the other genotypes.
In response to starvation for 48 hours at 8 or 9 weeks of age, the relative body weight loss of the Nuked Neck birds was not significantly different although it was slightly lower than that of nu+ nut birds, and the depression of body temperature and of plasma glucose did not differ from those of fully-feathered birds (Zein-el-Dein et ul., 1982).

Effects on plumage and on energy balance and its consequences
The effect of the Nu gene on plumage weight in proportion to live body weight in both the heterozygous and the homozygous states, causing an improvement of yield of the eviscerated carcase (Zein-el-Dein et ul., 1984b), is a direct consequence of the restriction by this gene of several pterylae and of feathers or the down cover of apteria. The effect itself has not been interpreted but it shows some analogies with that of the scaleless gene described by Abbot (1958) and others.
The insulating power of the plumage being well known, another expected consequence is an appreciable increase of heat loss, particularly at low or moderate environmental temperatures. This increase explains the higher feed intake and the poorer feed efficiency of Nuked Neck chicks and layers, especially the homozygotes, observed in experiments done at 15 to 20°C. The survival advantage of the same genotypes when submitted to a heat stress (Smith and Lee, 1977;Merat, unpublished) is likely to have the same origin. It is suggested (Monnet et ul., 1979) that the same cause may be responsible for the superior growth rate of Naked Neck (Nu Nu and Nu n u f ) chickens around 30°C and above. The heat production following meals, dissipated at a faster rate, would force the birds to reduce their feed intake less, resulting in a faster growth rate and at least as good feed efficiency as for normallyfeathered chicks. Bordas and Merat (1984) studying,data obtained at two different temperatures in a dwarf ( d w ) flock as compared with a normal ( D w ) flock, showed that the increase of the ratio of mean egg weight to body weight of the laying hen is a constant effect associated with the Nu gene in the homozygous state. The heterozygous genotype has a lower effect in the same direction. The excess of feed consumption of Nu Nu, and to a lesser extent that of Nu nu+ hens, in comparison with the nu+ nu' genotype, may contribute to this increase in egg weight by the supply of additional material, protein and so on, for egg formation. This hypothesis is 'not supported by direct evidence and there is no close relationship between the increase in egg weight associated with the N u gene and the corresponding increase of feed intake, according to ambient temperature or the size of the layer (Dw or d w ) . However, the effect of protein level on mean egg weight is known, as well as the fact that the optimum protein level for egg weight is higher than that corresponding to maximum egg number (Calet, 1972).

Effect on egg weight
On the other hand, data on average egg weight were published in the past for two other genes suppressing or reducing the extension of the plumage: sc (scaleless, autosomal recessive suppressing the plumage: Abbott, 1958;Abbott and Asmundson, 1962) and the K" allele at the sex linked K locus, which considerably delays feathering (Somes, 1975). In both cases, the mean egg weight was several grams higher in the presence of the mutant genotype without a parallel increase in body weight.

Depressing effect on body weight at low or moderate temperature
There is no satisfactory explanation for the substantial reduction in body weight of Naked Neck females (mainly homozygotes) at moderate temperatures (15 to 20°C) in the data presented by Monnet et al. (1980) and Bordas and Merat (1984). It is possible that the compensation for the energy requirement by additional feed intake associated with the Na Na genotype was not sufficient, as suggested from the slight depression of body temperature observed by Monnet et a f . (1980), and that in the long run the consequences on weight gain are cumulative. The fact that only the late growth period is concerned might be in accordance with this hypothesis, along with the fact that from 7 weeks of age the blood concentration of thyroid hormone T3 decreases, which may be associated with a decreasing metabolic rate (Decuypere et a f . , in press). On the other hand, the additional feed consumption associated with the Naked Neck type includes calcium. Can a difference in the consumption of this element have consequences on the rate of skeletal maturation? Ekerman e f u1. (1981) showed that lowering the ingested energy at 8 weeks of age increased the calcification of the tibia, but of course the situation is different from that associated with the Na vs. na+ gene.

Effects on meat yield and fattening
The interpretation of the relativity faster development of the muscular tissue associated with the Nu gene independently of body weight (Zein-El-Dein et al., 1984 b), especially in the thoracic region (El Attar and Merat. 1985), calls for further investigations. Results from Somes and Johnson (1980) pointing to 21 better breast conformation associated with the scaleless gene that causes a similar but more extreme reduction of the plumage than the Na gene, suggest that this may be a general effect associated with a reduction in the plumage. It has been indicated (Zein-El-Dein et uf., 1984 b) that during growth, the formation of a smaller amount of feathers leaves more protein material available for other tissues, if nitrogen retention is unchanged. On the other hand Goldspink (1977) indicated that the number of muscular fibres, which is one of the factors of further muscle growth, is determined during embryonic life. During this time, the distribution of an obviously limited total protein supply between down and muscles might be shifted in Naked Neck embryos to the advantage of muscular tissue.
Conversely, the slight reduction of the proportion o f subcutaneous and intermuscular fat in Nuked Neck birds (Zein-El-Dein et al., 1984 b) may correspond to the utilization of a higher part of the ingested lipids for thermoregulation.

Effects on semen traits and other effects
Among the other effects associated with the Nuked Neck phenotype, those concerning semen characters (Hammade et al., 1986 a & b), increase of volume and number of spermatozoa irrespective of the ambient temperature, are not explained at present, nor are the effects on behaviour traits noted by Zein-El-Dein et ul. (1984 c). The greater wattle size of Nu Nu and Nu nu+ birds is a local phenomenon. The extension of plumage around the wattles is reduced by the Nu gene, and this may be compared with the reduction of comb associated with crest (Cr gene), or with that of wattles in the presence of the Mb gene responsible for feather extension in this area (Hutt, 1949).

Potential value of the Naked Neck gene in poultry production
The most likely potential use of the Nuked Neck gene is at high ambient temperatures, from 25°C and mainly around 30°C and above.
The advantage of Nuked Neck broilers for growth (Table 2) is variable, increasing with temperature and possibly depending on other uncontrolled factors, but on the whole the use of the Nu gene appears as beneficial above 25"C, even if only the heterozygous Nu nut genotype is considered, which is more likely to be obtained in a terminal cross. Even at 25°C data (see later) show that Nuked Neck chickens are at no disadvantage. Further work however will be necessary to establish if the Nu Nu or Nu nu+ genotype has the same effect on body weight gains in a fluctuating temperature environment and to what extent it depends on relative humidity. On the other hand, the suggestion of Zein-El-Dein et ul. (1984 a) that Nuked Neck birds show a slightly stronger advantage in the presence of a ration with suboptimal protein level needs additional confirmation, possibly using a still lower protein ration than that used by these authors.
To this variable advantage for growth associated with the Nuked Neck genotypes at constant high temperature is to be added a consistent gain (about 1.5 to 2% with the heterozygous genotype) for yield of eviscerated or dressed carcase owing to the reduction of plumage. In addition there is a more variable but always positive effect on meat yield from 1 to more than 5 % ( Table 3) that mainly concerns readyto-cook chicken meat. The slightly lower percentage of subcutaneous and intermuscular fat in Nuked Neck chicks (Zein-El-Dein et ul., 1984 b;El Attar and Merat, 1985) is also favourable. The reduction of plerylae, and of the down between them, would probably result in faster plucking leaving fewer pinfeathers. This seems to correspond to the general opinion of producers, who do not claim any harmful effect on the frequency of breast blisters and carcase condition (Perrault, personal communciation).
Finally, a higher mean egg weight associated with Nu, if this allele is incorporated into a dam line, might have a slightly favourable effect on chick weight to 7-8 weeks of age, due to the influence generally found to be associated with this egg weight (e.g. Bray and Iton, 1962;Merritt and Gowe, 1964;Morris et al., 1968).
The depression of hatching rate suggested by Crawford (1977Crawford ( , 1978, Horst (1980) and our data (about 10% in relative terms) might be independent of climatic conditions. If so, it would be the only unfavourable effect associated with the Nuked Neck character in a hot environment. It affects the cost of the day-old chicks. Taking this cost as 14 to 16% of the total cost of the broiler (North, 1984), this effect would represent not more than 1.5% of the total cost, which is less than the amount of gains on body weight and yield.
In total, the gain on body weight alone expected from the incorporation of the Nu allele in a parental line would be equivalent to few generations of selection, possibly not more than 1 or 2, but this number would be distinctly higher taking into account the effects on yield.
For egg production at high temperature, the most constant advantage associated with the Nuked Neck genotypes is the increase of mean egg weight (Table 5 ) . For the Nu Nu genotype at 30°C the corresponding gain in a normal-sized (Dw) strain or in a strain fixed for the sex-linked dwarf gene ( d w ) amounts to 4.5 and 3.7 g respectively. For the heterozygous genotype in the same conditions, the gain is about halved, which is still appreciable. Additional results are needed to confirm the suggestion (Monnet et ul., 1979;Bordas and Merat, 1984) of a better shell strength and a higher laying rate after the first 5 months of laying for the Nuked Neck hens at 30°C.
The conclusions drawn for egg production are also likely to apply to female breeders in broiler production.
A t moderate temperature, the use of the Nuked Neck gene is obviously counterindicated for it appreciably reduces feed efficiency for both growth rate and egg production. Nevertheless, some results suggest that this defect may be corrected by a more precise control of the environment.
While at 15 to 20°C after the age of 4 weeks, Naked Neck chickens are clearly inferior for broiler production from the viewpoint of feed efficiency and sometimes for body weight at slaughter age (Table 2), it has been observed (unpublished data) that the performance of Nu nu+ and nu' nu+ chickens do not show any disadvantage of the former when raised at a constant temperature of 25°C f 1" from 4 weeks to a slaughter age (8 weeks). This trial concerned a total of 344 hatched chicks from a cross between Cornish males and Nu nu+ females from an experimental strain, the four groups corresponding to each sex and genotype being raised in separate identical floor pens. The mean 8-week average body weight for the Nu nu+ and nu+ nu+ genotype females was respectively 1407 and 1411 g, the corresponding feed efficiency from 10 to 56 days being 2.47 and 2.46. For Nu nu+ and nu+ nu+ males the mean 8-week weight was 1585 and 1598 g respectively, and feed efficiencies were respectively 2.34 and 2.33. None of these differences were significant. In these conditions, in the absence of any detectable disadvantage for the Nuked Neck bird with respect to body weight and feed utilization, the only expected difference of economic significance at the broiler stage is the slight but consistent advantage found previously (Table 3) for slaughter yield and meat yield for the Nu nut birds. This gain can be compared with the cost of additional heating, depending on climatic and economic conditio'ns and with the supposed loss in hatching rate associated with the Nu gene. For average conditions in France the cost of heating at 25°C for 4 weeks may be estimated grossly as representing between 1 and 1.5% of the selling price of the broiler (Stevens, personal communication). The loss from a poorer hatching rate of Nuked Neck chicks should be less than 1% of this price. Conversely, the gain from the improvement of yield of the eviscerated carcase can be estimated at a mean value of the order 1.8% and the more variable gain on meat yield is expected to be of the order 1 to 2% on the average. So it seems that in similar conditions to those in the experiments cited, the total balance is either close to zero or slightly positive, depending whether the effect on meat yield is taken into account or not. As suggested previously (Merat, 1981) this balance could be slightly improved if slaughter age is early so that the duration of additional heating is minimized.
For egg production, the data presented in table 5 show that when the temperature does not exceed 20°C, the effect associated with the Nu Nu and Nu nu+ genotypes is complex. Feed efficiency deteriorated at fixed body weight and production mainly for the homozygote. On the other hand, the weight of adult hens, especially for the Nu Nu genotype, is appreciably reduced and in spite of this the mean egg weight is slightly increased. The result on total feed efficiency expressed as g feed per g egg is clearly unfavourable for the Na Nu hens in a normal sized ( D w ) population. but possibly not for the Nu nu+ genotype. Moreover, it must be noticed that for dwarf ( d w ) layers the deterioration of residual feed efficiency (at fixed body weight and production) associated with the Nu gene is much more limited, so that the whole balance seems positive, or not unfavourable, at least for the Nu nat genotype (Bordas and Merat, 1984). Further work however should estimate in a more precise way the overall effect on feed efficiency in a dwarf stock and ascertain whether there is any harmful effect on shell strength for the Nuked Neck birds in temperate conditions. Conclusions More research is needed to establish in greater detail the environmental conditions in which the Nuked Neck gene may be beneficial, for either broiler or egg production. In both cases, the economic interest in breeding "Naked Neck" duplicates of commercial parental lines would depend on the extent of the possible area of use for the derived crosses. This breeding work, incorporating the gene into a line followed by repeated backcrossing to the commercial line, would be facilitated by the ease of identification of the homozygous and heterozygous genotypes for the Nu gene. Previously mentioned results indicate that, for broiler production a cross devised for hot climates, including one Naked Neck parent line, might also be used in temperate conditions with early slaughter and with the cost of additional heating to around 25°C. For egg production a dwarf Naked Neck layer for predominantly hot areas might combine the advantage of the Nu gene in these conditions and that associated with the dw gene according to Horst and Petersen (1979) and Merat et ul. (1974), but the use of dwurf Nuked Neck stocks for egg production in temperate conditions needs to be further investigated.
Such considerations will certainly not be decisive at present for the broiler industries in developed countries. They may deserve more attention in future in the context of specific climates and including egg traits and mortality but also if there were to be future plateauing of progress in the main selected traits.
The possible interpretation of these effects is discussed for each trait. Concerning the potential usefulness of the Naked Neck gene in poultry production, present-day results show on the whole an encouraging perspective at ambient high temperature, either for broiler or for egg production. At low or moderate temperatures an obvious defect associated with the Naked Neck gene is a deterioration of feed efficiency, either for growth or for laying, but the results cited suggest that this defect may be suppressed by a more precise control of the environmental temperature and possibly, as concerns egg production, by combining the Naked Neck and the sex-linked dwarf genes.