Comparative morphology of the larvae of the palm weevils Dynamis borassi (Fabricius) and Rhynchophorus palmarum (Linnaeus) (Curculionidae: Dryophthorinae): Two major pests of peach palms in the Neotropics

The morphology of the larvae of Dynamis borassi (Fabricius) and Rhynchophorus palmarum (Linnaeus) was compared, described and illustrated. A complete definition of subtribe Rhynchophorina at the larval stage was elaborated, based on mature and submature larvae, and characters of the first instar larvae are provided for the first time. Diagnostic characters to distinguish the larvae of D. borassi and R. palmarum are mainly from the epipharynx and maxillae, but the abdominal terminal plate also provides differential features that can aid in the identification. The characterization of both species at the larval stage is expected to be useful in the implementation of new control strategies in palm plantations in South America. For practical reasons, a comparative table is appended in Spanish to serve as a quick diagnostic guide for species identification.


INTRODUCTION
The weevils Dynamis borassi (Fabricius, 1801) and Rhynchophorus palmarum (Linnaeus, 1758) attack native, ornamental and commercially cultivated palms in tropical America (Couturier et al., 2000;Löhr et al., 2015), with significant economic impacts. Dynamis borassi has generated losses of up to 57% in experimental cultures of the palm Oenocarpus mapora Karst (Couturier et al., 2000). Meanwhile, for R. palmarum, its effect as a vector of the nematode Bursaphelenchus cocophilus (Cobb, 1919) on oil and coconut palms has been widely documented, having caused the elimination of large-scale production areas such as Tumaco Bay in Colombia (Löhr et al., 2015).
A similar situation has developed over the last decade in the production areas of peach palm, Bactris gasipaes Kunth (locally known as "chontaduro"), in Colombia. There has been a production loss of 48% (Agronet, 2016), attributable to the attacks of both D. borassi and R. palmarum (ICA, 2015;Pardo-Locarno et al., 2016). Although the presence of D. borassi has been documented, current management plans are designed exclusively to control R. palmarum (ICA, 2015). Consequently, peach palm production in most of the Pacific lowlands of Colombia has come to a stand-still (A.A. Vásquez-Ordóñez, unpublished). There is ample evidence now that the attack on the peach palm is initiated by D. borassi, and R. palmarum only comes in as an opportunist pest (A.A. Vásquez-Ordóñez, unpublished). This may be due to the difficulty to distinguish the larvae of both species, indispensable to determine the role of each species.
A key step in palm weevil research is to develop identification tools that allow field researchers to distinguish the larvae of these species since it is at this stage that major damage is caused. Morphological descriptions of the larvae of D. borassi and R. palmarum were made and illustrated by Wattanapongsiri (1966), in addition to those by Cotton (1924) and Anderson (1948). These authors recognize the high structural similarity of both species at the larval stage, highlighting the value of some mouthpart structures for differentiation. However, these larval descriptions differ in the terminology used and problems often arise when doing the identifications. Furthermore, details of body chaetotaxy and other taxonomic characters, widely documented in other weevils, are missing or poorly documented. Recently, Chamorro (2019) provided a synoptic key to larvae of genera of Dryophthorinae, based mainly on characters of the mouthparts. The key includes both genera, but information on Dynamis is rather incomplete as no material was available for examination, with only a few characters extracted from existing literature. Therefore, a comparative morphological study of the larvae of D. borassi and R. palmarum, using standardized terminology is urgently needed to facilitate accurate identification of both species at the larval stage. This is fundamental to document the occurrence and role of D. borassi in the recent wide-spread mortality of peach palms in Colombia. It is expected that the new morphological tools will facilitate rapid phytosanitary diagnoses in the field, since rearing larvae to the adult stage can take up to six months (Wattanapongsiri, 1966;Cuellar Palacios, 2019), making the monitoring and control processes too lengthy.

MATERIAL AND METHODS
We studied 23 larvae, 12 belonging to D. borassi and 11 to R. palmarum. All larvae of D. borassi were obtained from adults collected in pheromone traps (ferrugineol) from peach palm plantations in Colombia. The progenitors of five mature and four submature larvae were collected at Tadó, Chocó (05°17′08.82″N, 76°29′16.34″W) in March 2016. Two first instar and one second instar larvae were produced by females collected in Zabaletas and Bajo Calima, Buenaventura, Valle del Cauca (03°59′00.05″N, 76°58′49.77″W; 03°45′47.69″N, 76°57′49.69″W, respectively) in September 2017. The larvae were reared on artificial diet following the protocol described by Löhr (2016), in the entomology laboratory of the Colombian Corporation of Agricultural Research (AGROSAVIA), Palmira Research Center, Valle del Cauca, Colombia. On the other hand, a mature larva of R. palmarum was collected inside the stem of a peach palm at Pueblo Rico, Santa Cecilia (05°20′10.72″N, 76°08′23.06″W), in November 28, 2017. One first instar, four submature and five mature larvae of R. palmarum were reared from adults collected in pheromone traps at Palmira Research Center.
The taxonomic identity of the larval specimens and of their associated adults was corroborated with the keys of Wattanapongsiri (1966), Giblin-Davis et al. (2013) and Chamorro (2019).
The observations, photographs and illustrations for the morphological descriptions were done using the following equipment: Canon EOS 6D Mk ii camera with EF100 mm f/2.8 lens associated with a light box; Nikon DS-Ri1 U3 and JVC KY-F75U cameras, associated with a Nikon SMZ-1500 and a Leica MZ16f stereo microscopes, respectively. Subsequently, the larvae were dissected and prepared on microscopic slides as in May (1994) and Marvaldi (1999). Finally, the structural characters were documented and illustrated with a Leitz Laborlux S compound microscope and a Nikon SMZ800 stereomicroscope, using their attached drawing tubes. The terminology followed for body parts and setae of the larva of Curculionidae is explained in Marvaldi (1999), May (1994) and Oberprieler et al. (2014). Most larval specimens studied were deposited in the Entomology collection of Museum of La Plata, Buenos Aires, Argentina (MLP). The mature larvae of D. borassi, one first instar larva of R. palmarum, and adult specimens of both species were deposited in the Entomology Museum of the Universidad del Valle (MUSENUV), Cali, Valle del Cauca, Colombia.

RESULTS
The morphology of the larvae of Dynamis borassi and Rhynchophorus palmarum was compared, described and illustrated. A complete larval re-description was elaborated, based on mature and submature larvae, and a characterization of the first instar larva is newly provided. For D. borassi and R. palmarum, the diagnostic larval characters that distinguish them are given.

Subtribe Rhynchophorina (Figs. 1-9)
The following description applies to mature (and submature) larvae of both Dynamis borassi and Rhynchophorus palmarum, which are, respectively, the type species of the genera Dynamis and Rhynchophorus. It is descriptive of the larval stage of the subtribe Rhynchophorina of the current classification of dryophthorines (see Anderson & Marvaldi, 2014).

Comments:
The following additional adult characters (Wattanapongsiri, 1966;Löhr et al., 2015), in combination with those given above, distinguish R. palmarum from other species of Rhynchophorus: body entirely black, exceptionally reddish, interocular distance (dorsal view) ⅓ or less than ⅓ than the width of rostrum at base (Figs. 10G-H), rostral ventral area between the antennal scrobes rugose and with setae (Figs. 10K-L).

DISCUSSION
The larvae of Dynamis borassi and Rhynchophorus palmarum were redescribed in detail in this contribution, using modern terminology to allow comparisons with other larvae in Dryophthorinae and across Curculionidae. Few structural characters, including chaetotaxy, show clear differences between both species. In agreement with Anderson (1948) and Wattanapongsiri (1966), the most useful characters to distinguish D. borassi and R. palmarum are found in the mouthparts, particularly the epipharynx and maxillae. The observation of these features requires at least partial dissection of the head and examination under magnification, but the shape of the abdominal posterior plate, which also shows differences in both species, can be observed directly. This is advantageous to facilitate a rapid identification in the field, as required in most applied studies related to pest control. However, because differences in the abdominal plate of both species are only obvious and more distinct in older or mature larvae, it is worth to complement with other (e.g., from mouthparts) diagnostic features.
The terminology used by Wattanapongsiri (1966), to describe larval structures and chaetotaxy of Rhynchophorus and Dynamis, differs in some important aspects with the one adopted in the present work and other comparative studies (e.g., see Marvaldi, 1999;May, 1994;Chamorro & Huang, 2019). Among the differences worth to note are those involving the "anterolateral setae of epipharynx (als)", which in Wattanapongsiri (1966) are named "lateral setae of labrum" and, as additional source of confusion, their counting is given in full and not for one side, as it is standard for description of bilateral structures. Other discrepancies with Wattanapongsiri (1966) may arise because sometimes (e.g., in mandibles, stipes) he considers and counts as setae other sensilla that are not true setae (i.e., basal socket plus "hair"). In our study, we also provided labeled figures to facilitate the determination of homology when compared with works using different terminology.
From a phylogenetic perspective, the larval characters described for both species studied are in agreement with those diagnostic and synapomorphic of subfamily Dryophthorinae, such as maxillary malae with some dorsal setae branched, the pleura subdivided, the spiracles of AVIII placed on dorsum and oriented caudally, among others (May, 1994;Marvaldi, 1999; Anderson & Marvaldi,   Vásquez-Ordóñez, A.A. et al.: Larval morphology of Neotropical palm weevils Pap. Avulsos Zool., 2020;v.60.special-issue: e202060(s.i.).27 12/14 2014). We also corroborate, as diagnostic of dryophthorines, that the first instar larva lacks the spiracles of abdominal segments I-VII, having only the thoracic and AVIII spiracles developed. The larvae of later instars of Dynamis and Rhynchophorus have the complete set of abdominal spiracles, and although those of AI to AVII are small, they are clearly functional (i.e., connected to tracheal system, see Fig. 7H); therefore, they are not absent as interpreted in the identification key by Chamorro (2019). The occurrence of egg-bursters in first instar larvae of Curculionoidea is less common, but also more poorly documented, than in their sister group Chrysomeloidea, where egg burster's features (e.g., position, shape) were found to be phylogenetic informative at both higher and lower taxonomic levels (Cox, 1988). In Dryophthorinae, egg bursters were known to occur in first instar larvae of genus Sitophilus of subtribe Litosomina, and in the present contribution, we report their occurrence in representatives of Rhynchophorina.

ACKNOWLEDGMENTS
We are grateful to the different communities and secretaries of agriculture of the municipalities and departments visited, who facilitated the work. We also thank Inge Armbrecht and James Montoya, for their management of the Colciencias's project; Jackelinae Gaviria and Claudia Marcela Cuellar, for providing specimens; and Diana Marcela Urcuqui, Francisco Lopez Machado and Juan Felipe Ortega, for their technical help with photography at Universidad del Valle. This paper is dedicated in homage to Dr. Cleide Costa, in recognition of her contributions to the knowledge on immature stages and biology of Neotropical beetles.