Materials and Methods

Area of Study and Sampling. The area of study was established in the municipalities of Ayutla and Tecoanapa (Figure 1), in the spring-summer production cycle of 2011. Based on records from the nearest station (station 12012), the municipalities recorded an average annual temperature and rainfall of 27.7 °C and 1,519.9 mm (^{SMN, 2012}). For the selection of sites, stratified sampling was used (^{Ceja-Torres et al., 2000}), and the variation presented by the area with regard to altitude (masl) was considered. Four strata were defined as follows: stratum I) 100 to 300, stratum II) 301 to 500, stratum III) 501 to 700, and stratum IV) >701 (Table 1). Twelve producing locations were selected and in each one a sampling site was selected. The altitude and geographic position (latitude and longitude) of each site was determined using a GPS (Garmin Etrex^(r)) (Table 1; Figure 1). In each site selected, four samplings were carried out, directed at plants with "black leg" symptoms in the months of September, October, November, and December. Five plants were taken in each sampling, labeled for their identification, and taken to the laboratory for processing. In each site, the physical and chemical characteristics were determined in the Soil Physics Lab of the Colegio de Postgraduados. Texture was determined using the Bouyoucos hydrometer method, and pH (ratio 1:2 in H₂O) and organic matter, using ^{Walkley and Black's method (1934)}.

Figure 1 Sampling sites in the municipal areas of Ayutla and Tecoanapa, Gro. Mexico. Spring-summer 2011 production cycle. 

Table 1 Geografic ubication of samples sites in Ayutla and Teconapa, Guerrero, Mexico. 

Distribution and Frequency of Organisms. Fragments of tissue from diseased plants of approximately 0.5 cm were desinfected in a sodium hypochlorite solution at 1.5 % for 2 min. Later, they were rinsed three times with sterile distilled water, dried with sterilized paper towels, transferred to PDA culture medium, and incubated at 24±2 °C under controlled conditions for a week; the growth of each organism was then quantified. From the data obtained for each isolated organism for each sampling site, a distribution map was created. A means difference test (Tukey, p=0.05) was performed of the frequencies obtained from each species of microorganism (^{SAS Institute, 1988}).

Cultural and Morphological Characterization. The mycelial growths were transferred and purified using the monosporic culture and hyphal tip techniques in a water-agar medium at 2 %. In the preliminary identification, the colonies with cultural characteristics to Phytophthora were transferred on V8-agar culture medium for fourteen days, and in order to induce sporulation, disks, with a centimeter in diameter were retransferred in Petri dishes containing 25 mL of sterile distilled water. The remaining isolations were transferred on PDA culture medium. In all cases, the organisms were incubated once more at 24 ± 2 °C. The morphological isolation was carried out based on the taxonomical keys by ^{Holliday and Punithalingam (1970)}, ^{Booth (1971)}, ^{Mordue (1971)}, ^{Sutton (1980)}, ^{Hobbs et al. (1985)}, ^{Hanlin (1990)}, ^{Sutton (1992)}, ^{Erwin and Ribeiro (1996)}, ^{Leslie and Summerell (2006)}, ^{Barnett and Hunter (2006)}, ^{Gallegly and Hong (2008)}, and ^{Damm et al. (2009)}.

Molecular Characterization. DNA was extracted from the mycelia of the colonies of the different week-old isolations on PDA medium using the CTAB method (^{Murray and Thompson, 1980}) with some modifications (use of STE1x buffer and addition of PVP40 at 4 % to the extraction buffer). The mycelium of each species was macerated with 1 mL de STE1x buffer (Tris-HCl 100 mM pH8, EDTA 50 mM pH8, NaCl 100mM, β-mercaptoethanol 0.3 %). The samples were centrifuged at 14,000 rpm (revolutions per minute) for 8 min. The supernatant was decanted and 800 μL of extraction de buffer were added (Tris-HCl 100 mM pH8, NaCl 1.5 M, EDTA 20 mM pH8, CTAB 3 %, PVP40 4 %, β-mercaptoethanol 0.3 %). It was incubated for 40 min at 65 °C, and later, 700 μL of chloroform/isoamylalcohol were added (24:1); it was mixed and centrifuged for 8 min at 14,000 rpm. The supernatant was transferred to a new tube containing 800 μL of absolute ethanol. The DNA was precipitated for an hour at -20 °C, and then centrifuged for 8 min at 13,000 rpm. The supernatant was decanted and the precipitate was dried at 37 °C for 20 min to resuspend the DNA obtained in 100 μL of sterile ultrapure water. To amplify the internal transcribed spacer regions (ITS1-ITS2) and the intermediate region 5,8S of the rDNA of all the isolated organisms, except for Fusarium spp., the initiators ITS5 (5´GGAAGTAAAAGTCGTAACAAGG) and ITS4 (5´-TCCTCCGCTTATTGATATGC) (^{White et al., 1990}) were used. Meanwhile, for the species of Fusarium, the initiators ITS5 and NL4 (5´-GGTCCGTGTTTCAAGACGG-3´) were used, and to also amplify the dominions D1/D2 of the gene 28S (^{O'Donnell, 1993}). The reactions took place in a final volume of 25μL that contained 2.5 of 10X buffer, 1 μL of MgCl2 at 50mM, 0.5 μL of the mixture of dNTPs 10 mM, 2 μL of each initiator 10 pmol/μL and 0.2 μL of Taq-DNA polymerase 5U/μL (Invitrogen(r), USA). The reactions took place in a Biometra thermocycler, with the following program: initial denaturalization at 95 °C for 3 min, 35 denaturalization cycles at 95 °C for 30 sec (seconds), alignment at 55 °C for 45 sec and extension at 72 °C for 45 sec; a final extension of 72 °C for 7 min. The products of PCR were viewed by electrophoresis in agarose gel at 2 %. Once the presence of only one stripe was observed, the amplified DNA was sent to the company Macrogen (http://www.macrogen.com) to be sequenced.

A consensus sequences of each species was obtained by alignment (forward and reverse) using the software DNA Baser(r) ver. 4.16, and they were compared and aligned with those available in the database from the National Center for Biotechnology Information (NCBI) Gene Bank, U.S.A. using BLASTN 2.2.26 (^{Zhang et al., 2000}) and they were deposited in the Gene Bank.

Pathogenicity Tests. In October of 2011, each isolated pathogen was inoculated in three-month roselle plants; the plant material used was multiplied by "criolla" Guerrero botanical seeds, planted in pots with sterile soil and cultivated in a controlled environment chamber with a temperature of 28 °C during the day and 18 °C by night, with periods of light/darkness of 12 h. The Phytophthora parasitica inoculum increased on V8-agar medium for 14 days, and after this time, in order to induce the release of zoospores, discs with a diameter of 1cm were transferred into Petri dishes with 25 mL of sterile distilled water subjected to 5 °C for 5 min, and then to room temperature (24-27 °C) for 10-15 min; whereas for the increase in inoculum of the other isolated organisms, only Petri dishes with PDA were used. In this way, for Fusarium oxyporum, Fusarium incarnatum, Fusarium solani, Macrophomina phaseolina, and Colletotrichum truncatum growth was for 14 days, and 30 days for Phomopsis longicolla and Glomerella cingulata. The concentration of inoculum was estimated using a Neubauer chamber. For P. parasitica it was 2x10³ zoospores mL^-1. Conidial suspensions of F. oxyporum, F. solani, F. incarnatum, P. longicolla, and C. truncatum were inoculated at concentrations of 1x10⁶ conidia mL^-1. The inoculation of Macrophomina phaseolina was carried out with microsclerotia suspensions. In all cases, 10 mL of the suspension were inoculated per plant. For G. cingulata, ten PDA discs (1 cm diameter) were used, placed at the base of the stem of each plant. Five plants for each organism species (treatment) were inoculated and a control treatment with no inoculation was added.

A second test was carried out in the opencast, in the INIFAP- Campo Experimental Iguala, Guerrero in February of 2012, with average annual temperatures and rainfalls of 26.7 °C and 965.4 mm respectively (^{SMN, 2012}). For the increase of the inoculum, we once again used PDA and V8-agar medium. Five three-month old plants were inoculated for each species of fungi or oomycete (treatment) by inserting a sterilized toothpick impregnated with mycelium of the microorganism into the stem, plus a control without the inoculum. The plants under study remained opencast after being inoculated. When the symptoms induced by the inoculated pathogens presented themselves, these pathogens were reisolated using the technique described.

Results and Discussion

Distribution and Frequency of Organisms. From the plants collected and processed, five genuses of fungi and one oomycete were identified. Phytophthora parasitica (oomycete) was found in 10 sites, with frequencies that varied between 85 and 92.2 % (average of 72.4 %); Macrophomina phaseolina was isolated in four sites and in two of them, which were Tlachimala and Pozolapa, had high frequencies of 66.7 and 56.3 % respectively, and with an average of 10.6 %; Fusarium oxysporum was distributed in 11 sites with low isolation frequencies (3.6 to 18.8 %), and the average was 10.6%. The fungi Fusarium incarnatum, Phomopsis longicolla, Fusarium solani, Colletotricum truncatum, and Glomerella cingulata, with low frequencies (average of 0.4 to 2.4 %) and distribution (Figure 2, Table 2).

Figure 2 Distribution of organisms related to "black leg" of roselle during the spring-summer 2011 cycle in Ayutla and Tecoanapa, Guerrero. 

Table 2 Frecuency of aisolated organisms in roselle plants with "black leg" symptoms in 12 localities of Ayutla and Teconapa, Guerrero, México. Crop cicle Spring-Summer 2011. 

Table 2 shows that the frequency of Phytophthora parasitica was significantly higher (p=0.05); the other organisms showed no significant differences between groups, yet in terms of numbers, the above was followed by Macrophomina phaseolina and Fusarium oxysporum.

The diseased plants in which Phytophthora parasitica was most frequently isolated displayed generalized wilting, a yellow color, flabby leaves, necrosis in the base of the stem, which often extended up to the aerial section, reaching parts of the branches (Figure 3a and ^b ). Also, yet in a lesser degree, there were symptoms of necrosis observed along with cracking at the base of the stem with the production of exudates, which occasionally caused the flattening and death of the plant; these symptoms correspond to those reported by ^{Hernández and Romero (1990)}, and ^{Erwin and Ribeiro (1996)}. However, in diseased plants from Tlachimala and Pozolapa, where P. parasitica was not isolated, similar symptoms were found; in these places, necrosis was limited to the base of the plant's stem; plants were frequently found with detached epidermis and constriction in the area of progress of the disease (Figure 3 d). Plants displayed generalized wilting, a yellow color, flabby leaves (Figure 3c), and death of the plant. In these sites Macrophomina phaseolina was frequently isolated (Table 2). Locally, both symptomatologies are known as roselle "black leg".

Figure 3 Symptoms of "black leg" caused by different organisms. a) generalized wilting, yellowing, and leaf flabbiness; b) stem necrosis, extending to the branches, c) generalized wilting, yellowing, and leaf flabbiness. d) root and stem base rot with peeling of epidermis and constriction in the area of progress. 

This is the first report on M. phaseolina, F. incarnatum, F. solani, P. longicolla, C. truncatum, and G. cingulata, as organisms related to the "black leg" disease in this area of the state of Guerrero, in which we proved, based on pathogenicity tests, that not all were pathogenic.

On the other hand, the pH ranges varied from 4.7 to 6.9, organic matter content varied between 1.2 and 2.9 % and soil types were loam, sandy loam, and sandy clay loam. The organism Phytophthora parasitica was isolated from diseased plants in 10 sites which characteristically present soils with pH ranging from 4.7 to 5.5, whereas in two sites it was not isolated: Pozolapa and Tlachimala (Table 2) and presented a pH of 6.6 and 6.8, respectively (Table 3), sites located in stratum I. This indicates that the conditions of soils with pH close to neutral did not favor the presence of P. parasitica.

Table 3 Soil characteristics in plots of roselle (Hibiscus sabdariffa L.) in Ayutla and Tecoanapa, Guerrero, Mexico. 

pH= Potencial de Hidrógeno.

†MO=Materia orgánica (%).

There is evidence of soil pH being a critical factor for the formation of sporangia in some species of Phytophthora. In general, a high pH is toxic for the sporangia; in this sense, ^{Dasgupta et al. (2012)}, evaluated the behavior of P. parasitica in soils with a pH of 5.4, 7.0, and 8.5, and established that the athogen presented a better growth in a pH of 5.4. Likewise, ^{Besoain (2013)}, reported that the colonies of Phytophthora parasitica grow in a pH of between 5.5 and 6, and their growth was affected when it was higher than 6.5. In this study P. parasitica was isolated in soils with a pH of between 4.7 and 5.5, whereas in soils with a pH between 6.6 and 6.8 this organism was not isolated; these results are similar to those reported by ^{Dasgupta et al. (2012)} and ^{Besoaín (2013)}. On the other hand, in other crops, ^{Jha and Dubey (2000)}; ^{Surinder et al. (2013)}, determined that Macrophomina phaseolina prefers pH ranges between 6 and 7, similar to those found in Tlachimala and Pozolapa, where this pathogen was frequently isolated. No relation was found between the content of organic matter and texture (Table 3).

Cultural and Morphological Characterization. In the plants processed, eight species of microorganisms were identified: 1) Phytophthora pasitica Dastur developed dense colonies of a cotton-like growth in the shape of rosette, coenocytic mycelia, mostly presenting spider-like growth, predominantly oval-shaped sporangia, and with a prominent papilla; average measurements were 44.4 x 35.3 μm (length x width), intercalary and terminal chlamydospores (28 μm on average). These characteristics coincide with descriptions for this species by ^{Erwin and Ribeiro (1996)}, and ^{Gallegly and Hong (2008)}. 2) Macrophomina phaseolina (Tassi.) Goid. colonies displayed a gray color in their phase of growth and development, and became darker with age; they developed microsclerotia that varied in size (33-54 μm), round to irregular in shape with a black color, articulate and hard, septate mycelia; these characteristics for the species coincide with descriptions by ^{Holliday and Punithalingam (1970)}, ^{Abawi and Pastor-Corrales (1990)}, ^{Kaur et al. (2012)}. 3) Fusarium incarnatum (Robergge) Sacc. (=F. semitectum), initially white colonies with an orange pigment in the center, later became brown. In CLA (Carnation Leaf Agar) medium, it produced macroconidias with septa that varied in numbers from 3 to 5 (28.3-35.4 μm) and with a foot-shaped base cell, microconidias from 0 to 3 septa, these characteristics coincide with descriptions by ^{Nelson et al. (1983)} y ^{Leslie y Summerell (2006)} for this species. 4) Fusarium oxysporum Snyder and Hansen, pinkish-white colonies that turned violet with time, presented fusiformed conidias, generally with 3 septa measuring 23.3-49.9 x 3.1-5.2 μm, an abundance of microconidias, without septa and cylindrical, intercalary and terminal chlamydospores; these characteristics coincide with those reported by ^{Nelson et al. (1983)}, ^{Leslie and Summerell (2006)}. 5) Fusarium solani Snyder and Hansen displayed cream-colored colonies, scarce mycelia, abundant sporodochia present, frequently with a greenish pigmentation in the center, macroconidias between 5 and 7 septa measuring 28.3-41.5 x 4.3-6.4 μm, microconidias from 0 to 2 septa, conidiogenous cells in long monophialides; these characteristics coincide with the description by ^{Nelson et al. (1983)}, and ^{Leslie and Summerell (2006)}. 6) Glomerella cingulata (Stoneman) Spauld. and H. Schrenk, grayish-white to dark colonies, with asci that formed after 3 to 4 weeks, with nailed to cylindrical shapes (45.8-62.6 x 9.2-11.6 μm) slightly curved, with eight ascospores for every cylinder-shaped ascus (11.7-18.5 x 3.8-5.3 μm), unicellular and hyaline; this coincides with the cultural and morphologic description for this species by ^{Mordue (1971)}. (7) Colletotrichum truncatum (Schwein.) Andrus and W.D. Moore, flat colonies, grayish-white, they develop acervuli. Setae measured on average 80-150 μm, conidiophores were over 90 μm and conidia, slightly curved, measuring an average of 17.5-21.5 x 2.55-3.32 μm; these characteristics were consistent with the description of Colletotrichum truncatum (^{Damm et al., 2009}). 8) Phomopsis longicolla Hobbs (=Diaporthe longicolla) (^{Santos et al., 2011}), the colonies presented white, cottonlike growth, long pycnidia were observed which produced alpha-conidia and beta-conidia. The alpha-conidia were hyaline, ellipsoidal, to fusiform, measuring 4.05-7.57 x 1.48-3.25 μm. Beta-conidia, scarce, hyaline, threadlike, and with a hook-shaped tip, measured 18.2-34.5 x 1.3-2.9 μm. Cultural and morphological characteristics adjust to the description of Phomopsis longicolla reported by ^{Hobbs et al. (1985)}.

Molecular Characterization. The amplifications performed with the ITS5/ITS4 initiators amplified a fragment of approximately 900 pb for Phytophthora parasitica, and of approximately 600 pb for Macrophomina phaseolina, Glomerella cingulata, Colletotrichum truncatum, and Phomopsis longicolla. With the ITS5/NL4 initiators, a fragment of approximately 1100 pb was amplified in the Fusarium spp isolations. The consensus sequences of nucleotides obtained when compared to those available in the GenBank of the National Center for Biotechnology Information (NCBI), indicated a 99 % similarity (Table 4), which confirmed the identity of the isolated organisms at the species level. The consensus sequence of each microorganism was deposited in the GenBank (Table 4).

Table 4 Molecular characterization by alignment of sequences reported in the gene bank with the intergenic sequences (ITS ) of the rDNA genes of organisms isolated roselle plants with symptoms of "black leg". 

^¥NCBI (National Center of Biotechnology Information).

^ZAnamorfo.

Pathogenicity Tests. In the tests performed in a controlled environment chamber and in the opencast, Phytophthora parasitica turned out to be the organism with the highest pathogenicity, causing the deaths of 100 % of the inoculated plants. The plants that were inoculated with this organism in a controlled environment chamber showed symptoms of the disease 10 days after inoculation (dai) (Table 5). Opencast, the symptoms of the disease were observed 7 dai. Similar results were obtained by ^{Hernández and Romero (1990)} for P. parasitica. The fungus Macrophomina phaseolina, in a controlled environment chamber, produced symptoms in a plant at 28 dai (Table 5), and later, at an advanced stage of the disease (34 dai) formed microsclerotia in the base of the stem; whereas opencast, it produced symptoms in two plants at 13 dai (Table 5). In the sites where M. phaseolina was isolated in high frequently, with 66.7 % and 56.3 % (Tlachimala and Pozolapa, respectively), were the only ones in which the presence of Phytophthora parasitica was not detected, therefore we considered that M. phaseolina caused the symptoms observed. However, in Tepango and Tecoanapa, where M. phaseolina was also detected, though less frequently (4.2 % and 3.8 %) respectively, P. parasitica was isolated more frequently (Table 2), which is why the latter was considered to be the cause of the disease. In Mexico, up until the moment in which this investigation was carried out, there were no reports on the pathogenicity of M. phaseolina on roselle.

Table 5 Pathogenicity tests with organisms associated to "black leg" of roselle (Hibiscus sabdariffa L.). 

*1=Assay 1, tested in inviromental controled chamber. 2=Assay 2, tested on opencast.

Proportion of plants that showed symptoms on every assay.

Fusarium incarnatum caused symptoms of the disease only in opencast tests, in two of the five inoculated plants. The symptoms were observed 15 dai (Table 5), and characteristically included wilting and basal necrosis. Because this organism was isolated at a low frequency (2.4 %) (Table 2), it is considered by this study as a secondary pathogen of the "black leg" disease in the areas in which it was detected.

The results of the inoculations with Phytophthora parasítica indicated that this organism had the highest pathogenicity, followed by Macrophomina phaseolina, and the lowest pathogenicity was Fusarium incarnatum (Table 5). In Egypt, M. phaseolina and F. incarnatum were isolated from plants with symptoms of rooting rot and wilt of roselle. In artificial inoculations in greenhouses and in the opencast M. phaseolina was found to be more pathogenic than F. incarnatum (^{Hassan et al., 2014}), which coincides with the findings by this investigation.

From plants independently inoculated with P. parasitica, M. phaseolina, and F. incarnatum, and that produced symptoms, we obtained the same species that was inoculated when reisolating.

In the tests performed in a controlled environment chamber and opencast, as shown in Table 5, the other inoculated organisms and control plants did not show symptoms of the disease.

In a similar study carried out in Nigeria on roselle with stem and root rot, Phytophthora parasitica, Rhizoctonia solani, and Fusarium solani were isolated. In pathogenicity tests P. parasitica was found to be the main cause of the disease; R. solani only caused the death of plantlets, and F. solani was not pathogenic (^{Adeniji, 1970}).

Phytophthora parasitica has been reported in Guerrero by ^{Hernández and Romero (1990)} as being the cause of "black leg,", in India, it causes root and stem rotting in roselle (^{Kumar and Mandal, 2010}), and its presence has also been reported in Africa, Indonesia, Puerto Rico, Malaysia, Philippines, Ivory Coast, Brazil, and other countries, as the cause of root rot and stem in this species (^{Erwin and Ribeiro, 1996}; ^{Drenth and Guest, 2004}; ^{Silva et al., 2014}).

Likewise, in Cuba and El Salvador, ^{Wellman (1977)} reported that Macrophomina phaseolina causes stem rot in roselle plants, and recently, in Egypt, it was mentioned as inducing root rot and wilting of plants of this same species (^{Hassan et al., 2014}). In Bangladesh, it has been isolated from roselle seeds as a pathogen that affects germination (^{Islam et al., 2013}). It is also mentioned in other species of the genus Hibiscus. In India it causes "collar rot" in coral hibiscus (Hibiscus schizopetalus) (^{Santhakumari, 2002}).

In Egypt, the fungus Fusarium incarnatum (=F. semitectum) has been reported as the cause of root rot and roselle wilting (^{Hassan et al., 2014}). Also, in other countries, there are records of it being pathogenic in other species of the genus Hibiscus. For example, ^{Farr and Rossman (2015)} report F. incarnatum in kenaf (Hibiscus cannabinus L.) in Iran and Kenya, and on tulip (Hibiscus rosa-sinensis L.) in the Island of Barbados. ^{Leslie et al. (1990)} indicate that F. incarnatum is a common pathogenic agent in the soil, distributed mostly in tropical and subtropical regions.

The fungus Fusarium oxysporum was not considered pathogenic, despite reports of its pathogenicity in roselle in Malaysia, Nigeria, United States, and other countries (^{Ooi and Salleh, 1999}; ^{Amusa et al., 2005}; ^{Ploetz et al., 2007}; ^{Hassan et al., 2014}).

Additionally, in India Hibiscus sabdariffa is reported as a new host for Fusarium solani (^{Padaganur et al., 1988}). In Egypt, ^{Hassan et al. (2014)} mention that F. solani is a pathogenic agent for roselle, and ^{Chehri et al. (2014)} in Malaysia isolated it from roselle plants with symptoms of root rot. However, in this study it was not pathogenic.

The genus Colletotrichum includes several pathogenic species of plants with economic importance distributed mainly in tropical and subtropical regions (^{Cannon et al., 2012}). The fungus Colletotrichum truncatum is a pathogen of roselle in India (^{Farr and Rossman, 2015}), and it has also been reported in other species of the genus Hibiscus, such as H. esculentum in India and Pakistan or H. rosa-sinensis in Malaysia (^{Farr and Rossman, 2015}). Glomerella cingulata has also been reported on species of the genus Hibiscus such as H. tiliaceus as the cause of anthracnose in the United States (^{Farr and Rossman, 2015}) and on H. rosa-sinensis in Argentina (^{Rivera et al., 2000}). Both C. truncatum and G. cingulata have also been reported as endophytic species of the Malvaceae family, as well as other plant species (^{Costa et al., 2012}; ^{Kumar and Kaushik, 2013}). This coincides with findings from this study, in which they were not found to be pathogenic in roselle plants.

The genus Phomopsis is related to diseases in herbaceous plants, although it can attack bushes and some fruit trees (^{Agrios, 2005}). P. longicolla cause stem blight in peanut plants (Arachis hypogaea L.) (^{Sanogo and Etarock, 2009}); in soybean, it induces cankers in the stem, and it can infect pods and seeds (^{Lu et al., 2010}; ^{Li, 2010}). It has also been reported as being an endophyte in soybean and other species (^{Larran et al., 2002}; ^{Wagenaar and Clardy, 2001}; ^{Rhoden et al., 2012}). No reports were found on this organism as a pathogen of roselle (^{Farr and Rossman, 2015}), and in this study it was not found to be pathogenic.

Due to the above, the fungi F. oxysporum, F. solani, G. cingulata, C. truncatum, and P. longicolla are considered organisms associated to roselle and of low risk for this area of study, since they were not found to be pathogenic (Table 5). The frequency of isolation was low and in general terms, they displayed a limited distribution.

Conclusions

We conclude that Phytophthora parasitica is the organism with the greatest frequency and distribution in the region studied, and is therefore considered the main cause of the "black leg" disease. The fungus Macrophomina phaseolina presented low distribution, although it was isolated with high frequency in two sites, and therefore it is the cause of the disease in these places. Fusarium incarnatum was pathogenic, although it requires further study, due to its low frequency and distribution. Macrophomina phaseolina and Fusarium incarnatum are reported as pathogenic for roselle for the first time in Mexico. Fusarium oxysporum was found widely distributed, yet it presented a low frequency and it was not pathogenic. The fungi Fusarium solani, Glomerella cingulata, Colletotrichum truncatum, and Phomopsis longicolla are considered low risk for the crop, since they are not pathogenic and their isolation frequency was low.