INTRODUCTION
The white shrimp Penaeus schmittiBurkenroad, 1936 (Litopenaeus schmitti) (Pérez-Farfante & Kensley 1997) (Decapoda: Penaeidae) is distributed from the north coast of Cuba (23°N) to Guadaloupe (16°N) in the Caribbean Sea. On the Atlantic coast of Central and South America, this species is distributed from Cape Catoche (21°N), Mexico, to Rocha (34°S), Uruguay (Pérez-Farfante 1969, Rojas 1982, Zolessi & Philippi 1995, FAO 2016a). However, along with this extended distribution, the potential morphological differences interposed by physical barriers such as the Amazon River or the resurgence of Cabo Frio, Rio de Janeiro, Brazil, have not been studied enough. Fisheries of white shrimp P. schmitti are an important economic activity in several regions on the Atlantic coast (Neiva et al. 1971, García & Le Reste 1987, MPA 2011, Henriques et al. 2014, Calumby et al. 2016).
The species has already been subject to fishing in Cuban waters, reaching more than 800 t in 1983 (Sosa 2009). However, due to overfishing, reduced freshwater flow to the sea caused by river damming, and environmental deterioration, the shrimp fishery on the Cuban shelf has been reduced to minimum levels (Borrell et al. 2007, Sosa 2009, Espinosa et al. 2014). In order to increase these captures, actions were carried out in Cuba for the management of fishing, repopulation, and identification and improvement of the rearing zones (Páez et al. 1997). In other areas, P. schmitti fishery still produces relatively good catches. In Rio Janeiro, 3500 t were captured in 1986, and around 6000 t were landed in the northeastern Brazilian region in 2003 (Santos et al. 2005). According to data reported by the FAO, from 2011 to 2014, 4000 to 4400 t of this species were captured annually in Brazil under the common name of “southern white shrimp” (FAO 2016a,b). Around 8800 t of P. schmitti were captured by artisanal fishermen at Lake Maracaibo, Venezuela in 2007 (Alió et al. 2009, Andrade et al. 2009). Although the difficulty of having more accurate fishing statistics has already been pointed out (Prado & Neves 2015), the literature offers information on the catches of this species in many different places, which reflects the great fishing activity on this resource.
Commercial aquaculture of P. schmitti was practiced in several countries between 1980 and 2000. In Cuba, more than 2000 t yr−1 were produced by 2003 (Fonseca & Fernández de Alaiza 2003, FAO 2016a). In the initial development of P. schmitti farming in Brazil, it was considered the one species with the highest culture potential (Nascimento et al. 1991, Bezerra & Ribeiro-Alves 1995, De Paiva et al. 1995). Until the year 2000, the Pacific white-leg shrimp Penaeus vannamei was one of many shrimp species being cultured worldwide (FAO 2016b). Since then, shrimp farmers have increasingly preferred this species due to its zootechnical excellence (Ostrensky et al. 1998, Rönnbäck 2001, Peixoto et al. 2003). P. vannamei also has a greater resistance to certain diseases, such as infectious hypodermal and hematopoietic necrosis virus (IHHNV), which plagued the industry in the 1990s (Lightner 2003, Chamberlain 2010). For that reason, in most countries on the Atlantic coast of Central and South America, the commercial farming of P. schmitti and other native species has been completely abandoned and replaced by P. vannamei (Escobar 1985, Rosenberry 1990, Fernández de Alaiza et al. 2018).
The excessive dependence of the aquaculture industry on P. vannamei monoculture is not desirable by many different standards (ESPAE-ESPOL 2018). Furthermore, the use of native species in certain specific situations can contribute to the resilience of the activity (Henriques et al. 2014). Although zoo technically compared, P. schmitti is easily outperformed by P. vannamei; recent studies have confirmed the potential of P. schmitti culture (Marquez et al. 2012), especially using low stocking density and high protein level feed (Fraga et al. 2002). From an economic point of view, in certain niche markets, mainly those that value organic forms of production, native species may attain better prices, thus compensating for a less-than-ideal zootechnical performance. The cultivation of P. schmitti and pink shrimp Penaeus subtilis has been extensively developed in Guyana estuarine areas, utilizing wild-caught postlarvae (FAO 1997). The resistance to some diseases could be another advantage of native species. Laboratory tests have shown that P. vannamei’s susceptibility to white spot syndrome virus (WSSV) is significantly higher than P. schmitti (Unzueta-Bustamante et al. 2004). Many farms have completely ceased their activities in certain locations due to the continuous outbreaks of WSSV. More tolerant native species could easily compensate for their lower performance in these regions. Furthermore, P. schmitti farming could also help decline its fishing stocks (Rosenberry 2009, INSOPESCA 2013).
Taking this into account, the potential for cultivating P. schmitti in areas of the Atlantic coast of the Americas, where this species is considered native, brings some important advantages compared to P. vannamei. Firstly, it could contribute to the sustainability of shrimp farms (Henriques et al. 2014), especially those located at or near environmentally sensitive areas, such as conservation units. Another direct advantage is the possibility of culturing in open environments and the commercialization of juveniles as live bait, which is forbidden for exotic species. The use as live bait for sport fishing in the southeastern region of Brazil is considered a good business opportunity for small producers (AEN 2011, De Barros et al. 2014, Henriques et al. 2014).
Domestication and breeding of P. schmitti have already been the subject of research projects (Bécquer & Espinosa 2002, Bécquer et al. 2004, Pérez-Jar et al. 2007, 2010, Wilkenfeld et al. 2010). A future breeding selection program will have to rely on previous studies of morphological differentiation of populations of native species. Those projects need to establish the genetic diversity found among natural populations to plan crosses between these lineages (Da Silva 2007). Due to the great importance that P. schmitti farming had in Cuba, studies on the genetic variability of natural populations of this species were carried out with morphological techniques, allozymes, and microsatellites, targeting the improvement of captive broodstocks (Espinosa et al. 2001, 2002, Borrell et al. 2003, 2004, 2007). In Venezuela, García-Pinto (1970) studied morphometric characters of juveniles and adults of this species captured in Lake Maracaibo and the Gulf of Venezuela. This author found a high allometric growth by relating these characters to the cephalothorax length. Later, Andrade & Pérez (2004), studying the growth of P. schmitti from these same regions, proved that cephalothorax length was significantly different between the sexes. Other studies have been carried out on the northeastern coast of Brazil (Santos & Ivo 2004, Silva et al. 2018), targeting the growth, morphometric characterization, and mortality of P. schmitti in the natural environment and the population structure under strong fishing exploitation. In a study comparing P. schmitti, P. vannamei, and P. setiferus, Arena et al. (1997) determined that out of the 24 morphometric characters studied, the length of the cephalothorax (carapace), abdomen, and first abdominal segment were the most important variables, because they significantly differed between populations, species, and sexes. In contrast, in a detailed study of the morphometric characters of P. schmitti throughout its distribution, Pérez-Farfante (1969) found little geographic variation.
The study compared selected morphological structures in populations of P. schmitti occurring at the extremes of their natural distribution. In addition, morphological differences along the distribution affected by physical barriers were analyzed. It is reasonable to assume that, along with this distribution, differences between populations could be identified, reflecting the species’ phenotypic variability. Therefore, the objective of this study is to identify the morphological characteristics of different populations of P. schmitti, which could contribute to creating a baseline for the conservation of this native species on the Atlantic coast of Central and South America.
MATERIALS AND METHODS
From 2014 to 2016, samples of P. schmitti from the coastal areas of Cuba and Brazil were obtained by artisanal fishery or commercial trawling and by directly purchasing at commercial fishery landing points. All catching sites and fishing zones were georeferenced. Shrimp samplings were carried out in Manzanillo, Cuba (CUB, 20°39’N, 77°17’W) and six different points on the Brazilian coast: Touros, Rio Grande do Norte (RN, 05°20’S, 35°46’W), Aracaju, Sergipe (SE, 10°95’S, 37°01’W), Vitória, Espirito Santo (ES, 20°22’S, 40°15’W), Cabo Frio, Rio de Janeiro (RJ, 22°82’S, 41°90’W), Pontal do Paraná, Paraná (PR, 25°68’S, 48°39’W) and Laguna, Santa Catarina (SC, 28°40’S, 48°81’W) (Fig. 1). A total of 221 specimens were collected and immediately processed for morphological measurements. In addition, photographs of P. schmitti live specimens previously collected for other studies in Guaratuba Bay (PR, Brazil) and from Morrosquillo Gulf, Colombia, were considered. Lastly, another 559 specimens were obtained at Lucena, Paraíba (PB), Brazil (06°53’S, 34°52’W) to check color characteristics. These individuals were captured using beach-seine between November 2018 and May 2019.
Sampling areas were selected based on previous studies about the abundance of adults of P. schmitti (Neiva et al. 1971, De Loyola e Silva & Nakamura 1981, Andrade et al. 2004, Santos et al. 2008, Sosa 2009, Capparelli et al. 2012, Bochini et al. 2014), as well as contacts made with fishers and inhabitants of the sampling areas. At each location, at least 30 individuals were collected in similar numbers for females and males, mostly pre-adult and adult shrimps in the inter-molt stage. Individuals were kept in a styrofoam box with ice or in a refrigerator to avoid freezing or decomposition, affecting the measurements, especially on the appendices or soft parts (Lester 1983). All individuals were processed while fresh.
The order of shrimp measurements listed in Table 1 was followed during the data collection, starting with the whole animal and finishing with the measurement of abdominal length (Fig. 2). All measurements were performed by the same observer and always followed the same criteria (Barbosa-Saldaña et al. 2012). All individuals were measured using a digital caliper (Digimess, model 100.176BL ± 0.1 mm) and weighed using a precision digital scale portable (Bel Engineering, model M5-m241A ± 0.001 g). Before being weighed, excess moisture was removed from each specimen by wrapping it in a cloth for a few seconds. Next, the cephalothorax was removed from the abdomen manually (to obtain the abdomen weight with an exoskeleton); subsequently, the exoskeleton was removed to obtain the abdomen weight without the exoskeleton. Finally, the upper rostral teeth were counted with the help of a manual magnifying glass (Holden & Raitt 1975, Espinosa et al. 2002).
N° | Variable | MU | Description / Notes |
---|---|---|---|
1 | Partial total length (PTL) | mm | It is the distance between the posterior margin of the orbital cavity to the posterior edge of the sixth abdominal segment, with the abdomen fully extended (Fig. 2). |
2 | Total weight (TW) | g | Wet weight. |
3 | Sex (SX) | - | By observation of the reproductive organ (thelycum or petasma). |
4 | Maturity stage (MS) | - | According to Ramos & Primavera (1986). |
5 | Partial carapace length (PCL) | mm | Distance between the posterior end of the orbital cavity to the posterior border of the cephalothorax (Pérez-Farfante 1970b, Capparelli et al. 2012). |
6 | Carapace width (CW) | mm | The largest width, at the height of the last dorsal rostral tooth (Lester 1983). |
7 | First segment length (1SL) | mm | Distance from the posterior border of the cephalothorax to the posterior border of the first abdominal segment, with the abdomen, fully extended. |
8 | First segment depth (1SD) | mm | Depth at the midpoint of the first segment. |
9 | Fifth segment length (5SL) | mm | Distance from the posterior margin of the fourth segment to the posterior margin of the fifth, with the abdomen, fully extended. |
10 | Fifth segment depth (5SD) | mm | Depth at the midpoint of the fifth segment. |
11 | Sixth segment length (6SL) | mm | Distance from the posterior margin of the fifth segment to the posterior margin of the sixth, with the abdomen, fully extended. |
12 | Sixth segment depth (6SD) | mm | Depth at the midpoint of the sixth segment. |
13 | Abdomen length (AL) | mm | Distance between the anterior margin of the first abdominal segment and the posterior border of the sixth abdominal segment, with the abdomen fully extended. |
14 | Weight of abdomen with exoskeleton (WAE) | g | It is the wet weight of the abdomen, just separated from the cephalothorax. |
15 | Weight of abdomen without exoskeleton (WAW) | g | It is the wet weight of the abdomen, just separated from the cephalothorax and peeled. |
16 | Number of upper rostral teeth (NDR) | - | Counted using a hand magnifier (especially in smaller animals) without considering the last tooth (epigastric) (Pérez-Farfante 1969). |
MU: metric unit.
Additionally, digital images of the specimens were collected for each locality to document possible differences in external characteristics. The morphological measurement followed Lester (1983), with modifications proposed by Espinosa et al. (2002) (Table 1). For the measurement of partial total length (PTL) in this study, the rostrum and telson were not included since these structures often appeared broken (Nikolic & Ruiz de Quevedo 1970).
The frequency of occurrence related to the number of rostral teeth was calculated for all specimens, as it has been suggested to be important data on shrimp population studies (Espinosa et al. 2002). The correlation coefficient between the height of the sixth abdominal segment (H6S) and the weight of the abdomen with an exoskeleton (WAE) was also evaluated, as this measurement may be useful for weight estimations of live shrimp abdomen (Lester 1983). In addition, to compare all measurements recorded from different populations of the Atlantic coast, data were subjected to the coefficient of variation (CV) and a non-metric multidimensional scaling (nMDS) with Euclidean distances, following James & McCulloch (1990). In order to compare the same length classes among populations of different sampling points, a data filtering process was applied by restricting that at least four out of seven sampling points should be represented in every length class. This criterion was accomplished considering only shrimp between 75 and 124 mm of PTL. After this process, 199 data samples from 221 (90.5%) were finally used for the comparison analysis, which results are presented (Figs. 4-5).
The ANOVA analysis also compared different proportions regarding weight and length measurements (Barbosa-Saldaña et al. 2012).
RESULTS
Meristic variations
The frequency in the number of upper rostral teeth in the seven populations of Penaeus schmitti sampled in Cuba and Brazil, and frequency values found by Espinosa et al. (2002) are shown (Fig. 3). In Cuban and most Brazilian populations, this study found the number of rostral teeth ranging from 6 to 10, with a predominance (mode) of eight, except for SE and RJ populations, where organisms with nine rostral teeth prevailed. Among the samples of the P. schmitti population collected in the southeastern Cuban region of Manzanillo, a frequency of 93% of specimens with eight teeth was found.
Morphological variations
When analyzing the correlation between the sixth abdominal segment depth (6SD) and the WAE, a coefficient of r = 0.76 was found. However, a significantly higher correlation (r = 0.94) was found between the relationship between sixth segment length (6SL) and WAE.
The CV of all the measured variables in P. schmitti populations are shown (Fig. 4). It was considered appropriate to carry out this analysis, even with a relatively small sample size (n = 30), following the criteria described by Lester (1983) and Espinosa et al. (2002). The results of the ordination of shrimp measurements by nMDS are shown (Fig. 5). It can be observed that the greatest dispersion corresponds to data from CUB, RN, and ES, which are also the localities that presented the highest CV values. De la Fuente (2010) pointed out that larger animals show more differences between localities than smaller ones.
The average animal size for each locality (PTL) is shown in Table 2. Larger specimens corresponded to CUB, RN, and ES, showing higher data variation. On the other hand, the values corresponding to SE, RJ, PR, and SC show a greater similarity or clustering of the data of each location.
Studied populations | Mean PTL (mm) ± standard deviation | n |
---|---|---|
Manzanillo, Cuba | 100 ± 19.9 | 32 |
Touros, Rio Grande do Norte | 116 ± 20.0 | 32 |
Aracajú, Sergipe | 84 ± 5.3 | 32 |
Vitoria, Espirito Santo | 94 ± 14.5 | 32 |
Cabo Frio, Rio de Janeiro | 98 ± 6.5 | 31 |
Pontal do Paraná, Paraná | 106 ± 7.6 | 30 |
Laguna, Santa Catarina | 83 ± 4.7 | 32 |
The proportions of the main length and weight measures were also compared: abdomen length (AL)/partial total length (PTL), weight of abdomen with exoskeleton (WAE)/total weight (TW), and carapace width (CW)/partial carapace length (PCL).
The mean values calculated for these proportions are presented (Table 3). The shrimp samples collected in the southern Brazilian region of Laguna (SC), located almost at the limit of the southern distribution of the species, presented the longest (71.3%) proportionately and heavier tail (68.5%) among all populations compared. Conversely, the animals sampled in Manzanillo (CUB), at the northern border of the distribution, showed narrower mean carapace propor-tions (53.3%) and shorter abdomen (69.1%). Shrimp populations from ES and RJ presented the largest carapace width (57.0%). The shrimp from these last two locations did not present significant morphological differences, possibly reflecting geographic similarities.
Region | Country | Identified populations (From north to south) | Latitude | Proportions* (%) | Presence of white spots (in females, in the thoracic sternite XIV) | |||
---|---|---|---|---|---|---|---|---|
AL/PTL | WAE/TW | CW/PCL | ||||||
North | Cuba | G. de Batabanó | (1) | 22°N | 69.1a | Absent | ||
B. de Cienfuegos | (1) | 22°N | ||||||
Tunas de Zaza | (1) | 21°N | ||||||
Manzanillo | (1,2) | 20°N | 67.0ab | 53.3a | ||||
Central | Colombia | L. Navío Quebrado | (3) | 11°N | ||||
Cartagena-S. Marta | (3) | 11°N | Absent | |||||
G. de Morrosquillo | (3) | 9°N | ||||||
South | Brazil | São Luís, MA | (4) | 2°S | ||||
Camocim, CE | (4) | 3°S | ||||||
Fortaleza, CE | (4) | 4°S | ||||||
Diogo Lopes, RN | (5) | 5°S | ||||||
Touros, RN | (2,5) | 5°S | 70.2abc | 65.5c | 56.9ab | Present | ||
B. Formosa, RN | (5) | 6°S | ||||||
Lucena, PB | (2) | 6°S | Present | |||||
Recife, PE | (4) | 8°S | ||||||
Aracajú, SE | (2) | 11°S | 70.4b | 63.8a | 55.1ab | Present | ||
Vitoria, ES | (2,4) | 20°S | 70.7abc | 68.0bc | 57.0b | |||
Cabo Frio, RJ | (2) | 23°S | 69.6abc | 67.1c | 57.0ab | |||
Santos, SP | (4) | 24°S | ||||||
P. do Paraná, PR | (2) | 26°S | 70.6bc | 63.5a | 55.6b | Present | ||
Guaratuba, PR | (4) | 26°S | Present | |||||
Tijucas, SC | (4) | 27°S | ||||||
Laguna, SC | (2) | 28°S | 71.3ac | 68.5ab | 54.5b |
AL: abdomen length, PTL: partial total length, WAE: abdomen weight with the exoskeleton, TW: total weight, CW: carapace width, PCL: partial carapace length. Proportions tested using ANOVA and Tukey’s pairwise comparisons test in columns. Values with the same superscript letters do not differ. Populations previously studied by: (1) Borrell et al. (2004), (2) This study, (3) Valle et al. (2015), (4) Maggioni et al. (2003), and (5) Luvesuto (2006).
*Analysis of variance: proportion AL/PTL: F(6,192) = 4.01, P = 0.0008; proportion CW/PCL: F(6,192) = 4.02, P = 0.0008; proportion WAE/TW: F(6,192) = 14.77, P = 7.397.
MA: Maranhão. CE: Ceará, RN: Rio Norte, PB: Paraiba, PE: Pernambuco, SE: Sergipe, ES: Espirito Santo, RJ: Rio de Janeiro, PR: Paraná, SC: Santa Catarina.
Considering the data of all locations (n = 221), the total length-weight relationship is shown (Fig. 6). The calculated regression equation:
Additionally, for the estimation of the WAE using the 6SL and 6SD measurements, the equations WAE = 0.68680.1637 6SL and WAE = 1.45530.1592 6SD, respective-ly, were obtained.
Size variations
The literature reported the largest record sizes for natural populations of different Penaeus spp. species, including P. schmitti, are shown (Fig. 7). The larger sizes of shrimp collected for this study were observed in specimens captured in Manzanillo (CUB) and Touros (RN), with a total length/weight of 142 mm/62.0 g and 149 mm/70.4 g, respectively.
According to Díaz et al. (2014), the largest size (maximum total length) for P. schmitti is reported in Venezuela as 240 mm for females and 230 mm for males. In Figure 7, the asymptotic total length (TL∞) growth parameter was estimated for each species (Tabash & Palacios 1996, Andrade & Pérez 2004, Díaz et al. 2014).
Color variations
An interesting finding in this study was the presence of two whitish symmetrical spots located in the ventral thoracic region for females of P. schmitti collected in several localities of Brazil (Figs. 8a-b). These spots are in the last thoracic sternite (sternite XIV), with a shape and color that resembles the position and appearance of the spermatophore of males. These spots only appear in some of the females collected in Touros (RN), Lucena (PB), Aracajú (SE), as well as in Guaratuba and Pontal do Paraná (PR) (Table 3).
From the 559 individuals obtained at Lucena (PB), none of the 164 males presented the above-mentioned ventral spots. However, out of 395 females captured, 83% showed these spots. The females’ proportion with spots was less than 74% by date sampling. It was also worth mentioning that only the smallest and immature females (with less than 920 mm of maximum total length) showed the spots. Furthermore, these spots did not appear in the collections realized in Cuba nor in photographs of the specimens sampled in Colombia.
DISCUSSION
One of the main contributions of the present study was identifying a significant number of phenotypic differences among populations of white shrimp Penaeus schmitti from distant regions within the species’ natural distribution.
The meristic variations of the number of rostral teeth found in this study corroborated Pérez-Farfante (1969), who found a mode of eight rostral teeth in more than 300 specimens analyzed during her extensive research on the western Atlantic penaeid shrimps. However, the same author later found larger variations (5-10 teeth) when studying morphological characteristics of P. schmitti in different areas (Pérez-Farfante 1970b). Espinosa et al. (2002), studying P. schmitti in Manzanillo (CUB), reported a frequency of 64% of individuals with nine teeth. The authors also found differences in the frequency of upper rostral teeth between wild and captive P. schmitti, suggesting that, despite having a common origin, captive animals tended to show eight rostral teeth in opposition to nine found in wild populations. Contrarily, this work performed 12 years later in the same location could not find a single shrimp with nine rostral teeth.
The analysis of the correlation between 6SD and WAE, as suggested by Lester (1983), resulted in a coefficient significantly lower than reported by the author for P. vannamei (r = 0.95) and P. stylirostris (r = 0.85). However, a significantly higher correlation coefficient was found between 6SL and WAE for P. schmitti. As this measurement can be easily obtained with a digital caliper, the 6SL is recommended to estimate the WAE in P. schmitti specimens.
The proportion between abdomen and carapace may present significant variations within the same shrimp species and populations. Pérez-Farfante (1970b) reported that the carapace growth rate about total length growth is slower in P. schmitti juveniles, but this growth rate will increase when animals exceed 100-108 mm in total length. Although females of P. schmitti are usually larger than males, it is important to recognize that the proportion between carapace’s size and abdomen length can be distinct between sexes when comparing gross measurements within a species. Porto & Fonteles (1981), in a morphometric study of P. schmitti carried out in northern Brazil, found the male abdomen to be proportionally longer and heavier than the females. Conversely, García-Pinto (1970), working in Venezuela with this species, found two inflection points occurring when the carapace reaches 18.1 and 24.9 mm for females and 18.0 and 23.8 mm for males. The length of the abdomen was also used by Arena et al. (1997) to express significant differences among populations of the shrimp species Penaeus setiferus, P. vannamei, and P. schmitti.
Nikolic & Ruiz de Quevedo (1970) reported the collection of a specimen with 168 mm of PTL, using the same measuring method applied in the present study. The maximum absolute sizes observed in this study were similar to the results obtained by other authors for P. schmitti at the same Cuban location. The largest reported size of P. schmitti specimens in Brazil (186-202 mm) corresponded to Santos & Ivo (2004) collected between RN State and the mouth of the São Francisco River. On the other hand, results for partial total length in this work were up to 30% smaller than the report mentioned above because these authors included the rostrum and telson in the total length measurement. Certainly, in many studies, the length measurement included both or one of these structures (Pérez-Farfante 1970b, Araujo et al. 2009, Barbosa-Saldaña et al. 2012, Lutz et al. 2015). Although the lack of standardization in the measurement method makes it difficult to compare data from different references, revising the available literature makes it possible to infer from the maximum size reported for P. schmitti. This information could also be useful concerning the farming potential of P. schmitti.
The asymptotic total length is considered an indicator of P. schmitti potential growth, at least for the populations studied (Fig. 7). The maximum total length corresponds to the largest recorded specimen (mainly females) of Penaeus spp. according to Díaz et al. (2014) and FAO (2018a), caught in nature. Furthermore, the largest total length reported for a species of shrimp from the genus Penaeus corresponds to a specimen of P. schmitti. The fact that the maximum estimated size for this species is higher than P. vannamei is an indication that the growth potential of P. schmitti could theoretically be as high as P. vannamei (FAO 2018b), drawing attention to the possibilities for cultivating this native species.
There are several studies in the literature on the growth of P. schmitti in different locations in Venezuela and Brazil, which have identified populations with larger animals, differences in growth between sexes, and others (García-Pinto 1970, Andrade & Pérez 2004, 2007, Santos & Ivo 2004, 2006, Díaz et al. 2014). Western Venezuelan waters, especially the Gulf of Venezuela-Lake Maracaibo, have exceptionally favorable environmental conditions for the species. Historically, the largest shrimp catches, the densest populations (Pérez-Farfante 1969), and the larger sizes reported for P. schmitti correspond to these areas (Andrade & Pérez 2004, Andrade et al. 2009), which could be explained by the marked preference of P. schmitti for low salinity areas. This species is easily captured in river mouths and after heavy rains (Pérez-Farfante 1969, Nikolic & Ruiz de Quevedo 1970), and salinity in this region can seasonally reach values lower than 12 (Pérez-Farfante 1970b, Díaz et al. 2014). The available data suggests this area as a possible region for broodstock selection on a finer scale.
Regarding the whitish spots found in the ventral thoracic region of some Brazilian females of P. schmitti, it is interesting that there is no previous report of a similar feature in the literature. Even in the classic studies of Pérez-Farfante (1969, 1970b), which described the most frequent colors in the different stages of P. schmitti, there is no mention of these spots. Also, in personal contacts with different researchers in Cuba, no one declared to have observed such spots during years of previous studies on natural populations of the species. The reason this characteristic has not been previously mentioned may be related to the method of preservation. Pérez-Farfante (1970a) points out that certain features, like the exoskeleton coloration, can only be appreciated in fresh or recently preserved individuals. In the case of our work, the possibility of having live or freshly caught animals may have facilitated these observations.
For this reason, when analyzing the presence of these whitish spots, it is recommended that fresh specimens be used instead of preserved ones. The samples from Lucena (PB) show that this character seems to be associated with younger immature females. Could this be advantageous for them? Where and when this phenotypic character first appeared in females is an aspect that deserves a more detailed study. Never-theless, it could represent some kind of differentiation in the external characters among geographically distant populations of the species, possibly due to geographical barriers.
Geographical barriers
It is expected that distant populations present different degrees of differentiation due to their adaptation to their local environments. For instance, Barbosa-Saldaña et al. (2012) reported clinal morphological differentiation among four populations of the brown shrimp Penaeus californiensis, mainly related to the geographic distances along the Mexican Pacific coast.
Although, in the case of P. schmitti, differences between certain regions are much more due to the existence of localized environmental barriers than to a gradient of differentiation determined by a geographic distance. Luvesuto (2006), working with relatively close populations of P. schmitti on the coast of RN, found significant genetic and morphological varia-bility. This author suggested a possible influence of oceanic currents on the population structure, indicating the need for specific fishing management for each area to contribute to preserving natural stocks. This result agrees with Espinosa et al. (2002), Borrell et al. (2004), and Valle et al. (2015), regarding the existence of distinct populations of this species, even in relatively close localities and without geographical barriers of importance.
Genetic studies on P. schmitti mentioned the mouth of the Amazon River as an important barrier, which would have favored the divergence between the populations of the Brazilian and Caribbean geological provinces, especially during periods of eustatic sea-level decrease (Maggioni et al. 2001, Lastrucci 2011). Furthermore, Maggioni et al. (2003), using micro-satellite loci, found a high diversity in eight populations of P. schmitti from distinct localities along the Brazilian coast, with greater differentiation in the zone near Cabo Frio (23°S), where divergent ocean currents create a geographical barrier.
As the estuaries and mangroves inhabited by shrimp are considered among the most environmentally sensible ecosystems (Núñez-Solís 2013), it is to be hoped that by increasing the presence of this native species, the resilience of the ecosystem as a whole can be enhanced. Considering the extensive distribution zone of P. schmitti and the relatively low cost of morphometric studies, it is recommendable that such studies be extended to other regions of the Atlantic coast. Moreover, the preliminary knowledge about the populations in areas of greatest interest (such as the west coast of Venezuela, for example) could be further completed and confronted with genetic studies, which are more accurate but much more expensive. The methodology and knowledge obtained in the present study may be useful for the conservation of the natural populations of the species, considering that anatomical characteristics must be identified to avoid interbreeding in future restocking experiences and to direct cultivation efforts. These aspects must be considered when developing a technological package both for restocking and cultivation of P. schmitti, especially in environmentally sensitive areas.