O presente trabalho, com 122 dias de duração, foi realizado na Estação Experimental de Nova Odessa - SP, do Instituto de Zootecnia durante os meses de março a julho de 1967, com 24 leitões desmamados da raça Duroc-Jersey. Teve por objetivo estudar nas fases de crescimento, de terminação e de crescimento e de terminação o efeito de rações com diferentes níveis de proteína e de energia, no ganho médio diário de peso (GMD), nos consumos médios diários de ração (CMDR), de proteína bruta (CMD de PB), de nutrientes digestíveis totais (CMD de NDT), de energia bruta (CMD de EB), de energia digestível (CMD de ED) e de energia metabolizável (CMD de EM). Estudou-se também as conversões médias dirias de ração (CVMDR), de proteína bruta (CVMD de PB), de nutrientes digestíveis totais (CVMD de NDT), de energia bruta (CVMD de EB), de energia digestível (CVMD de ED) e de energia metabolizável (CVMD de EM). Considerou-se ainda o custo médio diário da alimentação (CMDA) e o CMDA por unidade de GMD. O delineamento experimental adotado foi de blocos ao acaso em esquema fatorial 2x3, com seis tratamentos e quatro repetições (duas de capadetes e duas de marrãs). Em cada tratamento considerou-se duas fases, em correspondência ao peso vivo, a inicial de crescimento (C) de 22,3 ± 0,4 kg até 59,3 ± 0,5 kg e a final de terminação (T) de 59,3 ± 0,5 kg até 104,4 ± 1,7 kg. Estudou-se também em conjunto os períodos de crescimento e de terminação (CT) de 22,3 ± 0,4 kg até 104,4 ± 1,7 kg. Admitiu-se no presente experimento os níveis de proteína e de energia, estabelecidos pelo NRC (1964) como normais, aqui designados de NP e NE respectivamente. Estes níveis normais foram aumentados em cerca de 10% para se obter rações de níveis alto de proteína (AP) e alto de energia (AE). De maneira semelhante, mas exclusivamente com respeito ao valor energético da dieta, fez-se uma redução de 10% a fim de obter rações de baixo nível de energia (BE). Ressalta-se que, tanto na fase de crescimento, como na de terminação, o nível em energia dentro de cada tratamento foi o mesmo, isso devido as necessidades em NDT, segundo ainda o NRC (1964), serem as mesmas para animais de 20 a 100 kg de peso vivo, o que não ocorre com o nível de proteína, o qual varia neste intervalo de peso com o próprio peso vivo. O esquema fatorial de 2x3 constou portanto de dois níveis de proteína em termos de proteína bruta (PB), normal (NP) com 16% e 13% de PB e alto (AP) com 17,5% e 14,5% de PB, sendo o primeiro nível para crescimento e o segundo para terminação. Os três níveis de energia, expressos em nutrientes digestíveis totais (NDT) foram baixo ( BE), normal (NE) e alto (AE) respectivamente com 67,5%, 75,0% e 82,5%. Estimou-se para todas as rações, através das suas fórmulas, os respectivos níveis de proteína bruta (% de PB), de nutrientes digestíveis totais (% de NDT), de energia bruta (kcal/kg de EB), de energia digestível (kcal/kg de ED) e de energia metabolizável (kcal/kg de EM). Através da análise bromatológica, foi possível determinar exclusivamente o nível de proteína em % de PB, o qual foi admitido como sendo o nível de proteína das rações estudadas. Utilizando-se a composição bromatológica dos alimentos, fornecida pelo NRC (1964) e NRC (1968), foi possível ajustar o nível de proteína, empregando o farelo de soja, enquanto que para o de energia recorreu-se à farinha de sabugo de milho restante fina e a graxa animal (banha comercial estabilizada). Calculou-se, para todas as dietas consideradas, as relações de energia/ proteína em tem os de % de NDT, kcal/kg de ED e kcal/kg de EM por % de PB. Relacionou-se também a% de PB do milho e a% de PB do farelo de soja em cada ração. Os animais permaneceram juntos numa área comum e foram alimentados à vontade com ração seca individualmente duas vezes ao dia, às 9 e às 15hs. As pesagens realizaram-se periodicamente em intervalos de 14 dias. Realizou-se a interpretação estatística dos resultados obtidos através da análise da variância, levando em consideração sempre o esquema fatorial. Através dos dados obtidos e analisados, por área investigada nesta pesquisa, puderam ser ressaltadas as seguintes conclusões estatisticamente significativas, a não ser o caso contrário, quando o resultado observado não revelou significância, nesta oportunidade então, após tais dados colocou-se entre parêntesis (o que não foi significativo).

The present experiment was carried out in the Nova Odessa Experimental Station of São Paulo State Secretary of Agriculture, during a 122 days period from March 1967 to July 1967, using 24 weaned Duroc - Jersey pigs. The object was to study in growing (G), from 22,3 ± 0,4 kg to 59,3 ± 0,5 kg, finishing (F) from 59,3 ± 0,5 kg to 104,4 ± 1,7 kg, and growing-finishing (GF) from 22,3 ± 0,4 kg to 104,4 ± 1,7 kg periods, the effects of different protein (P) and energy (E) levels on average daily gain (ADG) and including also feed (ADFI), crude protein (ADCPI), total digestible nutrient (ADTDNI), crude energy (ADCEI), digestible energy (ADDEI) and metabolizable energy (ADMEI) intakes. It was also studied average daily feed (ADFE), crude protein (ADCPE), total digestible nutrient (ADTDNE), crude energy (ADCEE), digestible energy (ADDEE) and metabolizable energy (ADMEE) efficiencies. It was also considered average daily feed costs (ADFC) and ADFC per kg of ADG. The experimental design was of randomized blocks in a 2 x 3 factorial scheme, with six treatments and four replicates, two with barrows and two with gilts. The 2 x 3 factorial outline consisted of two crude protein (CP) levels: normal (NP) as recommended by National Research Council (NRC), with 16% and 13% CP, and high protein (HP) with 17,5% and 14,5% CP, being for growing pigs the first level and for finishing pigs the second. The three energy levels were: low energy (LE), normal energy (NE), as also preconized by NRC, and high energy (HE), respectively with 67,5%, 75,0% and 82,5% TDN, in all the periods. It was estimated for all rations, based on their composition the respective levels of % CP, % TDN, and kcal/kg of crude energy (CE), digestible energy (DE) and metabolizable energy (ME). It was also determined, based on the chemical analysis, the level of CP. The protein level was been adjusted using soybean oil meal while the energy level was adjusted using finely ground corn cobs or stabilized lard. It was calculated for all diets the E/P relationships and CP from com and CP from soybean oil meal, COP/SCP. The animals were fed individually, twice a day, with dry feed “ad libitum”, while the weighing took place every 14 days. The data were interpreted by analysis of variance, always considering the factorial scheme. In the present trial pigs that received HP rations, with larger amounts of soybean oil meal, did not perform as well as those fed with NP, as recommended by NRC. This trend became more evident in ADG and in feed and distinct nutrient intake. The performance was particularly poor when it was in the concomitant presence of HP and HE in the G phase. The analyzed efficiencies revealed better results for NP in G phase, while in F phase detached the HP. Because of this, including the G and F phases, there would be compensation between opposite results, which at last, showed evidences that were low, irregular and non-significant. It can be rebounded that HP level revealed almost always higher intakes and lower efficiencies of crude protein. Although in some dietetic conditions the HP level can give lower feeding costs, in general this protein level was responsible by the higher feeding costa per kg of ADG. Because of this the method of increasing swine carcass meat percentage through the use of diets with high protein did not seem very satisfactory especially when considering the financial results. In swine production with corn-soybean oil meal rations a more adequate relation of CCP/SCP seems to be more important than the quantity and/or quality of dietary protein. That relation apparently must be near 1:1 in the G period and somewhat higher than 1 for the corn in the F period. On the other hand, employment of rations with LE, including corn cobs relatively to NE recommended also by NRC, evidenced that such energetic reduction can present lower ADG, smaller ADTDNI, including also the ADDEI and ADMEI. Moreover it was verified that LE was responsible by lower ADFE, ADCPE and ADCEE, although in particular conditions it can present smaller ADFC but not ADFC per kg of ADG. The diets with LE was employed with the same goal of obtaining carcasses with greater meat percentage. Although it could have the advantage of resulting in lower ADFC, it presents the disadvantage that, perhaps, part of protein can be used as a energy source, according to what attested the verified lower ADCPE. On the other hand, the use of diets with HE containing stabilized lard resulted in the finding of higher ADG and lower ADFI and ADCPI. It was distinguished; however, most lower ADG for HE level when coincided with the presence of HP, in G period. The rations with HE almost always were better with respect to ADFE, ADCPE and of the different forms of energy. Nevertheless this HE level had a serious inconvenient to increase too much either ADFC or ADFC per kg of ADG. It seems that the quantity of energy ingested or necessary for a kg of gain is more important than the level of energy in the diet, like suggests the NRC. This hypothesis has a support in the fact that those diets with a variation of protein and energy did not affect statistically either ADTDNE or ADMEE in all together G and F periods. It was also noted that diets with certain protein and energy levels, like HPHE presented ADTDNE and ADMEE practically equal in the G, F and GF phases. It was observed that E/P relations of the diets had effects on various characteristics studied. Thus it was noted that relations wider than those recommended by NRC, resulted in higher ADG. This fact apparently strange perhaps could be explained by the fact that the swine’s in this experiment were not typically the “meat type” hogs. Really the expected results should be opposite, since, under tropical conditions, the energetic requirements for swine maintenance are lower than in temperate zonas, for which the recommendations of NRC were elaborated. It is interesting to point out that it was not possible to identify any significant effect due to swine sex on the studied characteristics.