Body dimension comparisons between slick and wild type-haired Puerto Rican Holsteins 1,2

A previous study from our group evaluating growth in dairy heifers suggests that Puerto Rican Holstein slick-haired heifers (SLICK) could reach maturity earlier and at smaller body size than their wild type-haired (WT) counterparts. Thus, the present study aimed to determine if such differences exist in mature cows by comparing several body dimensions (i.e., body weight, withers height, hip height, thoracic perimeter, barrel, and shoulder to pin bone distance) of 24 SLICK and 54 WT lactating Holstein cows at the Agricultural Experiment Station in Lajas, Puerto Rico. Also, body weight was divided by each dimension to obtain the respective ratios. Data were analyzed by the GLIMMIX procedure of SAS. The SLICK cows presented larger barrels (236.54±2.68 vs. 231.06±2.29 cm; P=0.0363) and body weight / shoulder to pin bone distance (3.51±0.09 vs. 3.32±0.08 kg/cm; P=0.0218) than their WT counterparts. However, the shoulder to pin bone distance was smaller in SLICK than in WT cows (180.07±1.82 vs. 185.20±1.56 cm; P=0.0059). No significant differences between hair coat groups were observed in any other evaluated variable. The SLICK cows had shorter and deeper bodies than WT cows. Future studies should evaluate if such differences could impact their productive efficiencies. obtener las respectivas relaciones. Los datos se analizaron mediante el procedimiento GLIMMIX en SAS. Las vacas PELONAS presentaron mayores barriles (236.54±2.68 vs. 231.06±2.29 cm; P=0.0363) y relación peso corporal / distancia entre el hombro y la punta del anca (3.51±0.09 vs. 3.32±0.08 kg/ cm; P=0.0218) que las vacas REGULARES. Sin embargo, la distancia entre el hombro y la punta del anca resultó menor en las vacas PELONAS que en las vacas REGULARES (180.07±1.82 vs. 185.20±1.56 cm; P=0.0059). No se observaron diferencias significativas entre grupos de pelaje en las demás variables evaluadas. Las vacas PELONAS presentaron cuerpos más cortos y profundos que las vacas REGULARES. Estudios futuros deben evaluar si la eficiencia productiva puede ser afectada por estas diferencias. Palabras de pelo corto, peso corporal, dimensiones


INTRODUCTION
Puerto Rico's dairy industry exclusively uses temperate Bos taurus cattle breeds (Sánchez, 2018), which are highly susceptible to heat stress (Silanikove, 2000). These animals are normally longhaired (wild type-haired; WT). Fortunately, a short-haired phenotype (SLICK), highly adaptive to the local tropical climate, also exists in the major dairy breeds in Puerto Rico. Due to the negative impact that heat stress exerts on the productivity of temperate Bos taurus cattle (West, 2003), SLICK cattle received considerable scientific attention in recent years because of its superior thermoregulatory capacity. Studies have determined that, when exposed to heat stress, mature SLICK cows maintain lower body temperatures  and respiration rates , have larger sweat glands (Contreras-Correa et al., 2017;Muñiz-Cruz et al., 2018) and higher milk production values .
In a recent study from our group (Muñiz-Cruz et al., 2017) we observed differences between the growth trends of SLICK and WT Puerto Rican Holstein heifers from four to 34 months of age, suggesting that the first group reaches maturity earlier and with a smaller body size. This may be an important finding because studies of WT Holstein cattle from Minnesota have associated a smaller body size with superior productive efficiency, relative to similar large sized animals (Mahoney et al., 1986;Yerex et al., 1988;Hansen et al., 1999). To this author's understanding, however, body size-related comparisons between mature SLICK and WT cows in Puerto Rico are not available in literature. Thus, the current study aimed to determine if such differences in body dimensions in fact exist between mature SLICK and WT Puerto Rican Holstein cows.

MATERIALS AND METHODS
Body dimensions were compared between 24 SLICK and 54 WT lactating Puerto Rican Holstein cows from the Agricultural Experiment Station dairy herd in Lajas, Puerto Rico. Animals in the lactating herd were considered as mature cows. The SLICK and WT groups were balanced for milk yield (16.78±1.16 and 16.92±0.68 kg/d, respectively; P=0.9189), parity (2.39±0.30 and 2.06±0.20 lactations, respectively; P=0.3660) and days in milk (195.39±27.01 and 211.66±18.32, respectively; P=0.6197). Hair coat types were first phenotypically classified and then genomically confirmed. The author visually classified as phenotypically SLICK those cows with a short, sometimes glossy hair coat all around the body; while the cows with the typical longer Holstein hair coat were considered WT. After following the procedures previously described by Littlejohn et al. (2014), Dr. Melvin Pagán (University of Puerto Rico at Mayagüez) generously provided the genomic classifications for all animals. Cows with a visually intermediate hair coat length were not considered. Cows with contradictory phenotypic and genomic classifications were not evaluated. The day after body weight was recorded by an electronic scale, all cows were restricted in headlocks in the farm's free-stall barn (for daily concentrate feeding), and the following measurements ( Figure 1) were obtained: withers height, hip height, thoracic perimeter, barrel, and shoulder to pin bone dis- tance. Also, the cow's body weight was divided over the respective body dimensions in order to obtain the specific ratios. The withers and hip height were determined by a Measuring Stick (Nasco; Ft. Atkinson, WI) 4 and the thoracic perimeter, barrel, and shoulder to pin bone distance were recorded by a commercial fiberglass measuring tape (Ace Hardware Corporation, Yauco, PR). All measurements were obtained with the cows standing straight and square on a level surface. During sampling, concentrate feed was offered at the headlock feeder. The dependent variables (body weight, withers height, hip height, thoracic perimeter, barrel, shoulder to pin bone distance, and the ratios of body weight over each body dimension) were evaluated by the GLIMMIX procedure of SAS. Hair coat type was included as a fixed effect in the model, while the cow's identification number was considered the random effect. Significant differences were detected at a P<0.05. Table 1 presents the body dimension comparisons between hair coat groups. The SLICK cows presented larger barrels (236.54±2.68 vs. 231.06±2.29 cm; P=0.0363) and body weight / shoulder to pin bone distance (3.51±0.09 vs. 3.32±0.08 kg/cm; P=0.0218) than their WT relatives. However, the opposite was observed in the shoulder to pin bone distance, where SLICK cows were smaller than their WT counterparts (180.07±1.82 vs. 185.20±1.56 cm; P=0.0059). No significant differences between hair coat groups were observed in any other of the evaluated variables.

RESULTS AND DISCUSSION
Although not evaluated in the present study, the observed differences in barrel circumference may be explained by differences in feed intake between hair coat groups. It is well known that heat stress limits feed intake in dairy cattle (West, 2003). However, because SLICK cows have a superior thermoregulatory capacity Contreras-Correa et al., 2016;Sánchez-Rodríguez et al., 2016;Contreras-Correa et al., 2017;Muñiz-Cruz et al., 2018), a smaller impact on feed intake may be reasonably expected in this group when exposed to tropical weather conditions. The diet of the evaluated cows consisted of limited access to concentrate feed immediately before milking and ad libitum access to tropical grasses the remaining daily period. Therefore, different intakes of concentrate feed are unlikely between hair coat groups because the feed is offered in limited quantities under the shade of the barn [which limits heat stress impact, and the first milking occurs early in the morning (0300 h) when there is no solar radiation] and cows normally show a marked preference for this kind of food. However, when cows are taken to the pasture paddocks (where solar radiation is a major environmental issue in the tropics) it is not uncommon to observe SLICK cows grazing under the sun during the hottest hours of the day; while all their WT relatives rest in the shade, lying in the mud in an attempt to alleviate the negative impacts of heat stress (Sánchez-Rodriquez, 2019). In fact, such differences in intake may help to explain the greater milk production reported in SLICK cows by Delgado et al. (2014) and Contreras-Correa et al. (2016). In order to eat greater amounts of tropical grasses, a larger ruminal capacity is required, since tropical grasses are high in fiber (Stobbs, 1995), including lignin, which is especially high during the dry season (Detmann et al., 2009). In fact, it has been reported that the addition of fiber to the diet of Holstein calves (Khan et al., 2012) and cows (Hale et al., 1940;Dado and Allen, 1995) increases their barrel circumferences and ruminal capacities. Such high lignin content limits fiber digestion in ruminants (Jung, 1987), probably making necessary greater intake to achieve the aforementioned superior milk production. Therefore, a larger dry matter intake of high fiber tropical forages may help to explain the observed differences in barrel dimensions as well as the lack of any difference in body weight between hair coat groups (P=0.4247; Table 1), as a larger barrel may add weight and compensate for the shorter body observed in SLICK cows.
If the proposed effect of a larger ruminal capacity and barrel on the body weight of SLICK cows could be excluded, the observed shorter bodies in this group may indicate differences in mature body size between phenotypes. A previous study from our group (Muñiz-Cruz et al., 2017) observed that from four to 34 months of age, while the WT Puerto Rican Holstein heifers grow in a linear pattern, their SLICK relatives presented a quadratic relation between age and body weight. Normal growth curves in cattle increase at a fast, close to linear rate during the juvenile stage, begin to stabilize as maturity approaches, and remain steady with no further changes during adult life (McDaniel and Legates, 1965;Berry et al., 2005). Thus, in the Muñiz-Cruz et al. (2017) study, by 34 months of age the SLICK heifers may have been approaching the stabilization stage of the curve, while their WT counterparts were still in the fast growing, juvenile stage. Therefore, these results suggested that SLICK heifers may be able to reach maturity earlier and with a smaller body size than the WT ones, supporting the findings of the present study. In fact, Bergmanns (1847; as reviewed by Meiri and Dayan, 2003) stated that animals from the same species present a smaller body size when they adapt to the tropics, in comparison to those raised in temperate locations. He explained that a smaller body size implies a larger surface area to body volume ratio, which facilitate heat dissipation under hot environmental conditions. Concurrently, the larger body weight / shoulder to pin bone distance ratio observed in the SLICK cows reflects the previously discussed lack of difference in body weight between hair coat groups and a shorter body.
Having shorter bodies and greater feed intake may result in productive advantages, which support the aforementioned greater milk yields in SLICK than in WT cows (Delgado et al., 2014;. Since 1966, the University of Minnesota has selected their Holstein cattle in terms of body size, resulting in both small and large sized herds (Mahoney et al., 1986;Yerex et al., 1988;Hansen et al., 1999). When compared, the small sized herd presented lower disease incidence (Mahoney et al., 1986), longer productive lives (Hansen et al., 1999), and greater feed efficiencies (Yerex et al., 1988) than their large-sized counterparts. The concept of "dilution of maintenance effect" establishes that if two cows with similar body weights, but different milk productions, are compared, the one with the larger volume of milk will be more efficient in terms of energy use (VandeHaar and St-Pierre, 2006;Capper et al., 2009;Bauman and Capper, 2010). That is so because cows with similar body sizes will have similar energy requirements for maintenance. But even though a greater milk production implies a greater energy requirement for milk synthesis, the total energy costs per unit of milk produced will be diluted. As previously mentioned, if the assumption of a greater feed intake in SLICK cows is correct, removing the ruminal content may result in a smaller body weight in the SLICK cows. Therefore, the energy requirements for maintenance may be smaller in the SLICK group, since body size directly influences the energy requirements for maintenance (Demment and Van Soest, 1985;Veerkamp, 1988). Taking this into consideration, SLICK cows may be more efficient by both a smaller maintenance requirement and a greater volume of milk produced.
In order to maintain the greater productivity normally reported in SLICK cows (Delgado et al., 2014;Contreras et al., 2016), both a greater feed intake and a greater degree of ruminal fermentation are required. Greater ruminal fermentation has been associated with a larger amount of metabolic heat produced and greater body temperatures in cattle (Russell, 2007). However, several studies from our group have reported lower body temperatures in the SLICK than in the WT cows (Sánchez-Rodríguez et al., 2015;Sánchez-Rodríguez et al., 2016). There may be two possible explanations for this apparent discrepancy. First, the SLICK cows have achieved a series of important adapta-tions to tropical weather, including a shorter hair coat (Sánchez, 2019) and larger sweat glands (Contreras-Correa et al., 2017;Muñiz-Cruz et al., 2018), that allow for greater heat dissipation in comparison with WT similar cows. Second, if the author's observations are correct, the SLICK group only eats larger quantities of tropical grasses, but not of concentrate feed. This is important in terms of ruminal heat production because the non-structural carbohydrates (starch) in concentrate feed are easily fermentable in the rumen, allowing large quantities of heat to be released (Alzahal et al., 2011). However, the structural carbohydrates found in forages are slowly and partially fermented in the rumen, resulting in less heat production (Alzahal et al., 2011).

CONCLUSIONS
Puerto Rican SLICK cows have a shorter and deeper body in comparison with their WT counterparts. These results are consistent with previous findings from our group where SLICK heifers appeared to reach maturity at a smaller body size than the WTs. In the literature there are examples of the multiple productive advantages that small cows have over larger cows. Consequently, future studies should evaluate if such differences in growth trends and body size could affect the animal's productivity in a tropical scenario.