Several species of cactophagous moths exist in North America (Neunzig 1997), including the invasive Cactoblastis cactorum (Berg) and the native Melitara prodenialis Walker. Both of these latter 2 species infest members of Opuntia, where they co-occur in the southeastern United States (Baker & Stiling 2009). Because C. cactorum is rapidly spreading and poses a threat to rare Opuntia species in Florida and the deserts of North America (Stiling & Moon 2001, Marsico et al. 2011), efforts have been made to control the dispersal of C. cactorum. One method of C. cactorum control, utilized in South Africa and suggested for the United States, involves collecting and destroying eggs from impacted cladodes (Zimmermann et al. 2000). Though egg removal is not currently a strategy employed for C. cactorum control in the U.S., all research efforts aimed at controlling damage and spread of C. cactorum require the collection of moths for laboratory experimentation and rearing, and often include the collection of eggs. Moreover, increasingly the similar, native, co-occurring M. prodenialis is being studied in comparative experiments with C. cactorum (e.g., Woodard et al. 2012). Therefore, it is important for researchers to be able to distinguish eggs laid by these 2 species.

Cactoblastis cactorum and M. prodenialis are now sympatric and select the same host plants (Baker & Stiling 2009). Therefore, it is possible that using indiscriminant techniques to remove eggs and cladodes will result in the destruction of native M. prodenialis individuals, negatively impacting their populations. As a best practice, invasive species management should negatively impact native species as little as possible. Additionally, resources should not be spent removing moths or moth-damaged plants from populations comprised entirely of M. prodenialis. As such, it is necessary to distinguish between the 2 insect species at all life stages. Larvae can easily be identified at middle and late instars based on their color; however, identification of early instar larvae is more difficult. Another means of identification at any life-stage is the sequence obtained from the COI gene (Marsico et al. 2011; Sauby et al. 2012), though this is impractical for rapid laboratory or field use. Here we report a quantifiable method that trained researchers and invasive species managers can use to distinguish between the eggsticks of C. cactorum and M. prodenialis.

Both species lay eggs one on top of the other to form eggsticks that are superficially similar in appearance (Hubbard 1895; Hoffmann & Zimmermann 1989). Close inspection of C. cactorum and M. prodenialis eggsticks indicate distinct differences between the two. Despite needs for a reliable method of distinguishing between the eggsticks of these species, differences have not been quantified (Rose et al. 2011). To make the distinction between moth species, photographs of field-collected C. cactorum and M. prodenialis eggsticks were used to measure several aspects of the eggstick morphology.

TABLE 1.

COMPARISON BETWEEN CHARACTERISTICS OF CACTOBLASTIS CACTORUM AND MELITARA PRODENIALIS EGGSTICKS.

MATERIALS AND METHODS

Eggsticks of C. cactorum were collected from 5 locations in the Florida Panhandle, U.S.A., in May 2009, Jul 2009, and May 2010. Eggsticks of M. prodenialis were collected from 4 locations in the Florida Panhandle, U.S.A., in May 2009, Jun 2009, and May 2010. Measurements were taken on photographs of the field-collected eggsticks using Adobe Photoshop CS5. Photographs of 163 C. cactorum and 141 M. prodenialis eggsticks were taken with a metric ruler included as a size reference. For each eggstick, the number of eggs was counted, and the lengths and widths of the terminal and middle eggs were measured using the Photoshop CS5 measuring tool. Egg length was calculated in the same direction of eggstick length, and egg width was calculated perpendicular to length. If the eggstick was straight, the measure tool was used to directly take the length of the entire eggstick. For eggsticks that were curved, a path was created with the pen tool down the length of the stick. The path was then filled with text (i.e., periods), which was copied and pasted on a straight path and measured with the measuring tool. Curvature was measured as change in angle per change in unit length. The length of the sharpest curve was taken, and the angle of that curve was taken by the measuring tool. A Student's t-test was used to test for significant differences between species for number of eggs, terminal egg length, and terminal egg width. Due to a violation of assumptions of the parametric test, middle egg length and width, eggstick length, and curvature were analyzed using a Mann-Whitney U-test.

RESULTS

Qualitatively, there are differences in egg color between species as the eggs reach maturity. Cactoblastis cactorum eggs become gray, silvery, and darken to black as the embryos mature. Melitara prodenialis eggs become increasingly darker brown as the embryos mature. There were significant differences between the species (P < 0.001 for Student's t or Mann-Whitney U tests) in the average and ranked values for all the characteristics measured (Table 1; Fig. 1; Fig. 2). Based on data associated with variable means, in addition to data ranges, the following dichotomous key for species delineation was developed using interquartile values.

Fig. 1.

Eggsticks belonging to Cactoblastis cactorum (a) and Melitara prodenialis (b). For both species, darker eggsticks are more mature, and lighter eggsticks are less developed. Cactoblastis cactorum eggs that are close to hatching are almost black (see for example the 4th eggstick from the right in panel a). Melitara prodenialis eggsticks also darken as they near hatching (see for example the 5th eggstick from the right in panel b). A tight curve at the basal end of the longer, thinner C. cactorum eggsticks is typical, while M. prodenialis sticks are usually straighter.

Fig. 2.

Boxplots comparing characteristics of Cactoblastis cactorum and Melitara prodenialis eggsticks. Boxplots show the means (horizontal lines within boxes), interquartile values (horizontal boundaries of the boxes that represent 25% and 75% of values that fall within the boxes), and the ranges (minimum and maximum values depicted as hatches at the end of vertical lines) for each variable measured, (a) Number of eggs, (b) terminal egg length, (c) terminal egg width, (d) middle egg length, (e) middle egg width, (f) eggstick length, and (g) eggstick curvature, measured as change in angle over length (i.e., °/mm). All lengths were measured in millimeters.

DISCUSSION

Useful characteristics for identification are those that are distinct with minimal overlap between species. Even though there are significant differences for all variables measured for the 2 cactus moth species, the range for all variables overlaps between species (Table 1). For this reason, a single character cannot be used as the sole identifier, and therefore, the characteristics need to be considered together to make reliable positive identifications. Because the length and width of eggs in C. cactorum and M. prodenialis differ significantly by only fractions of a millimeter, such small differences are not easy to discern outside the laboratory. Yet with practice, individual egg sizes can be distinguished by the observer. Cactoblastis cactorum and M. prodenialis eggsticks differ significantly in length and number of eggs, both which can be measured and counted easily (Fig. 1). Degree of curvature, particularly for long eggsticks and particularly positioned at the base of the eggstick, also tends to be greater in C. cactorum, which adds an important characteristic for identification purposes (Fig. 1).

We have now quantified the observable differences for the important diagnostic characteristics of eggsticks of C. cactorum and M. prodenialis. As an important invasive species with a large impact, it is critical to be able to distinguish C. cactorum from the similar, native, co-occurring species M. prodenialis, particularly at phases of the life cycle that are poorly characterized. Additionally, because these organisms are the focus of current and future research, scientists will benefit from the results reported here. Though the quantified differences and dichotomous key presented are useful for scientists and other individuals experienced with this study system and invasive species management system, the similarities between the eggsticks of these moths may make field identification impractical for invasive species control programs based on C. cactorum eggstick removal. These programs often rely on volunteers and assistants that may not have the time to become experienced in nuanced identification of the more difficult life stages of these species, and invasive species managers would not want inexperienced workers to leave C. cactorum eggsticks in the field based on the mistaken identification. Still, prudence in invasive species control strategies must be exercised with respect to similar native species, and populations of M. prodenialis that do not contain C. cactorum should not be put at risk of removal. We hope that utilizing these characteristics will add to the practical arsenal used by invasive species managers in their attempts to minimize invasive species impact of C. cactorum while simultaneously working to protect native populations of M. prodenialis.