A biological switching valve evolved in the female of a sex-role reversed cave insect to receive multiple seminal packages.

We report a functional switching valve within the female genitalia of the Brazilian cave insect Neotrogla. The valve complex is composed of two plate-like sclerites, a closure element, and in-and-outflow canals. Females have a penis-like intromittent organ to coercively anchor males and obtain voluminous semen. The semen is packed in a capsule, whose formation is initiated by seminal injection. It is not only used for fertilization but also consumed by the female as nutrition. The valve complex has two slots for insemination so that Neotrogla can continue mating while the first slot is occupied. In conjunction with the female penis, this switching valve is a morphological novelty enabling females to compete for seminal gifts in their nutrient-poor cave habitats through long copulation times and multiple seminal injections. The evolution of this switching valve may have been a prerequisite for the reversal of the intromittent organ in Neotrogla.

Introduction

Many man-made engineering solutions have evolved already in insects. Such examples include hinges (flapping flight enabled by the wing base: Brodsky, 1994), on-off valves (spiracle openings to regulate airflow: Chapman, 1998), backflow valves (the bombardier beetle's defensive spray: Arndt et al., 2015), coiling mechanisms (genital tubes: Matsumura et al., 2017a) or catapult-like mechanisms (the legs of many jumping insects: Burrows, 2013). Some mechanisms that were formerly thought to be unique to human engineering have been discovered recently in insects: biological screws have been found in a beetle's leg (van de Kamp et al., 2011), and interacting gears were found in the jumping legs of planthoppers (Burrows and Sutton, 2013). The design and construction of such mechanisms on a micrometer scale is a challenging task in engineering (Feinberg et al., 2001). Therefore, studies of micron-scale biological structures can be rewarding as they illuminate construction principles in insects that could be applied to technical solutions in engineering (Matsumura et al., 2017b).

The genus Neotrogla (family Prionoglarididae) is a minute Brazilian cave insect belonging to the order Psocodea (booklice, barklice, and parasitic lice). This genus is of special evolutionary and morphological interest because of the reversal in its genital structures (Yoshizawa et al., 2014). The females of Neotrogla have a penis-like intromittent organ (gynosome: Figure 1A), which is inserted to a male vagina-like genital cavity for copulation. During mating, the male injects liquid semen into the female’s sperm storage organ (spermatheca) through the opening of the spermathcal duct at the tip of the female penis. Within the spermatheca, the injected semen then induces the formation of a hard capsule shell around itself (Figure 1A,B: Wearing-Wilde (1995); Yoshizawa et al., 2014). Although there is only a single inlet spermathecal duct present, occasionally two seminal capsules are attached simultaneously to a plate-like structure on the spermatheca (termed ‘spermathecal plate’: [Lienhard et al., 1893; Yoshizawa et al., 2014]).

Figure 1.

Morphology of the spermatheca and spermathecal plate of Neotrogla.

(A) Whole abdomen of a virgin female showing the location of the mating system. T8 and 9 indicate tergites 8 and 9. (B) Light microscopy photograph of the spermathecal plate with a single seminal capsule. (C) CLSM image of the spermathecal plate. (D) 3D segmentation of the spermathecal plate with no seminal capsule. Dotted line indicates the section plane shown in F and G. (E) 3D segmentation of the spermathecal plate with two seminal capsules and showing seminal flow. (F) Attachment points for the seminal capsules in the virgin female. (G) The base of two seminal capsules in different stages of connection to the spermathecal plate (corresponding to the two seminal capsules in Figure 1E).

The semen within the capsule is used not only for fertilization but is also consumed by the female as nutrition. To compete for nutritious semen, the direction of sexual selection is reversed in Neotrogla (sex-role reversal: Yoshizawa et al., 2014). Each seminal capsule is voluminous (~0.05 mm3, corresponding to ~300 ml scaled up to humans), and the duration of the copulation is very long (for 40 to 70 hr). In a closely related species lacking reversal of genital structure (Lepinotus patruelis, Trogiidae), the seminal transfer for forming a similarly voluminous seminal capsule is known to complete in 50 min (Wearing-Wilde, 1995). The female penis of Neotrogla bears a lot of spines, by which females anchor a male coercively during copulation. Therefore, females are obviously responsible for this very long copulation, probably to obtain more semen from a male (Yoshizawa et al., 2014). After consumption of the semen, the empty capsule is detached from the plate, which is kept within the spermatheca. Because females frequently have empty capsules within the spermathecal pouch (in an extreme case, up to nine empty capsules and two filled ones attached to the plate were observed: Yoshizawa et al., 2014), female Neotrogla can be considered polyandrous, which is apparently controlled actively by the female.

Although several examples of sex-role reversed animals induced by seminal gifting have been reported, Neotrogla so far is the only example with a morphological reversal of its intromittent organs (Kamimura and Yoshizawa, 2017). This morphological reversal seems to require further upstream modifications of the genital system, such as the formation of the spermathecal plate, which was reported as an additional novelty tightly associated with the seminal gifting (Lienhard et al., 1893; Yoshizawa et al., 2014). However, the detailed morphology and function of this spermathecal plate, which presumably plays an important role in the evolution of the female penis, remained unclear due to the extremely small size of the involved subcomponents and their fragile spatial composition.

In this study, we investigated the structure of the spermathecal plate by using a combination of confocal laser scanning microscopy (CLSM) and high-resolution synchrotron microcomputed tomography (HR-µCT) to assess the functional morphology of sperm storage and control of seminal flow. We examined three species of Neotrogla (N. brasiliensis, N. aurora and N. truncata), the spermathecal morphology of which is practically identical (Lienhard et al., 1893). Based on the results, we discuss the evolutionary significance of this plate during the evolution of the reversed intromittent organs in Neotrogla.

Results

The spermatheca of Neotrogla is in principle composed of an extensible pouch for storage of the seminal capsule, a spermathecal plate divided into two interconnected sclerites (body sclerites 1 and 2: Sc1 and 2 hereafter), and the spermathecal duct, which discharges into the spermathecal plate (Figure 1CD, Video 1). The terminology used for the following description is summarized in Table 1.

Video 1.

3D segmentation of the spermatheca and two seminal capsules attached to the spermathecal plate (see Figure 1E).

Table 1.

Valve terminology used in the text

Actuator	Device used to operate a valve using electric, pneumatic or hydraulic means
Body	The principal pressure-containing part of a valve in which the closure element and seats are located
Closure element (CE)	The moving part of a valve, positioned in the flow stream, that controls the flow through the valve, for example wedge, plug, clapper, ball
Controller	A device that directs the flow of a valve
End connection	The type of connection supplied on the ends of a valve that allows it to be connected to piping — may be a welded end, flanged end, threaded or socket weld
Pennation angle	The oblique attachment of single muscle fascicles to the CE. It was measured as the angle between the outermost fascicles in a given muscle bundle

The switching valve mechanism is located at the spermathecal plate. Sc 2 is a bowl-shaped chitinous structure harboring a fan-like muscle (the actuator), which originates at its ventral part and attaches to a thumb-shaped controller/closure element (CE: Figures 1CD and 2A–D). Given their attachment area, the force range of each muscle bundle is between 0.0758–0.6562 mN (assuming a standard intrinsic muscle force of 33 N/cm²: David et al., 2016). The CE is located at the dorsal connection of Sc1 and 2, where the spermathecal duct opens into the pouch (Figure 2A–D).

Figure 2.

Morphology of the closure element (CE) and its associated structures.

(A) Muscle attachment to CE. (B) 3D segmentation of Figure 2A. (C) Detail of CE. (D) CLSM image of CE and neighboring structures. (E) Schematic illustration of the function of the switching valve system in closed (left) and opened (right) conditions.

In the virgin female, Sc 2 is connected to Sc 1 laterally through an interlock-like structure composed of several ridges on both sides that fit into each other (Figure 1F). This is the location where seminal capsules are formed during copulation (Figure 1G). Both Sc1 and Sc2 possess a system of patches of resilin, a rubber-like protein found in arthropod cuticles wherever potential energy is stored for spring-like motions or bending of structures. The region surrounding the CE contains extensive resilin patches (Figures 1CD and 2). Two larger patches are located where the seminal capsules are attached so that this region can expand to harbor the base of the seminal capsule (Figure 1DF). Another resilin-rich region is located around the valve mechanism (Figures 1CD and 2), where the resilin serves to passively keep the valve opening in the closed position. Opening of the valve can only occur through activation of the actuator muscle bundle. In the fixed material analyzed, the muscle bundles are contracted due to the fixation process. Therefore, the switching valve is in an opened position in Figures 1–2.

The seminal fluid entering the main lumen of the duct can be directed into the left or right channel depending on the position of the CE. If the left part of the actuator muscle is activated, the CE is moved to the left, so that a channel on the right side opens, which then allows for seminal flow into the right capsule (Figure 2E). A small lateral extension at the distal end of the CE serves to close the opposite channel during the opening of the other channel (Figure 2DE). If the right part of the actuator is activated, the process is executed in reverse. In a female fixed during copula, seminal flow from the duct opening lumen toward one of two seminal capsules was clearly observed (Figure 1E). Seminal flow can thus be directed by the female through differential muscle activation that moves the CE.

Measurements of each structure and the estimated power produced by the discernible muscle bundles mentioned above are summarized in Table 2.

Table 2.

Measurements of each component of the spermathecal plate.

Object	Pennation angle	Attachment area of muscle [cm2]	Muscle strength [mN]	Length [mm]	Volume [µm3]	Mass [µg]
Muscle m1	34.16	0.0000199	0.6562	0.1
Muscle m2	30.18	0.0000191	0.6319	0.08
Muscle m3	15.59	0.0000023	0.0758	0.09
Muscle m4	44.55	0.0000133	0.4374	0.09
Muscle m5	12.26	0.0000033	0.1096	0.09
Muscle m6	32.75	0.0000073	0.2394	0.11
Sum of muscle strength			2.1503
Closure element				0.03
Spermathecal pouch					1847443779	0.0021246
Body sclerite 1					199709300	0.0002297
Body sclerite 2					101434410	0.0001166

Discussion

The present analyses show that females of Neotrogla use a special mechanism to actively control the direction of seminal flow (Figures 1E and 2A–D). This biological switching valve allows females to receive two seminal packages (i.e., more nutrition) from the same or different males within a short time span. In particular, the switching valve allows to receive a second seminal capsule while the first one is consumed. A similar but less sclerotized structure can be observed in close relatives of Neotrogla (Sensitibilla and Afrotrogla, all belonging to the tribe Sensitibillini), suggesting that the structure probably originated in their common ancestor (Lienhard, 1893; Lienhard et al., 1893). Neotrogla is distributed in South America, whereas Sensitibilla and Afrotrogla are distributed in southern Africa. Therefore, the origin of this switching valve dates to at least the break-up of the two continents, over 100 million years ago (Seton et al., 2012). In other psocodeans, no sclerite or muscle corresponding to those of the spermathecal plate elements have been observed (Badonnel, 1934; Klier, 1956; Wearing-Wilde, 1995). The spermathecal plate thus clearly represents an evolutionary novelty (Müller and Wagner, 2003).

The mechanism of flow control and redirection within this biological switching valve is fundamentally different from that in man-made switching valves, which are used, for example in the oil and gas industry. Technical switching valves use the active rotation of tube or ball elements with various openings to redirect the flow and both closing as well as opening involve an active movement of the valve. In contrast, the observed biological switching valve involves a passive movement component. The seminal flow injected by a male is redirected by a differential inclination of the CE (Figure 2E). This inclination is due to the fan-like geometry of the actuator muscle whose parts can pull the CE in the preferred direction to work against the passive closing forces generated by the resilin patches around the valve complex. This muscle-closure element configuration could be advantageous since each muscle bundle pulls in an optimal direction to incline the CE into one of the two opening positions. Additionally, there is no need for lubrication since the moveable parts do not move against each other. Compared to technical switching valves, this design is advantageous to some extent because the muscle forces are not redirected via a lever arm.

The presence of a spermathecal plate with a valve function to control seminal flow may further refine our understanding of the causes and consequences of the reversed sexual selection in this genus. Female-female competition for males (sex-role reversal) to receive nutritious seminal substances is considered as the most important factor driving the evolution of the female penis (Yoshizawa et al., 2014). Among all animals with known sex-role reversal, Neotrogla is the only example in which a female penis evolved (Kamimura and Yoshizawa, 2017). In a close relative of Neotrogla, the barklouse species Lepinotus patruelis (Wearing-Wilde, 1996), the sex-roles are also reversed, but the species possesses normal genital structures. Therefore, it is very likely that, in addition to the sex-role reversal, there is at least one other key factor that enabled the evolution of a female penis. Although females of all three known genera of Sensitibillini possess a spermathecal plate, females of Sensitibilla do not have a penis-like organ (Lienhard, 1893; Lienhard et al., 1893). This strongly suggests that the evolution of the spermathecal plate, possibly including a switching valve, preceded the evolution of the female penis.

The spermathcal plate has two slots available for insemination (Figure 1E). In Neotrogla and the related species (Lepinotus), content of the capsule is digested as nutrition during the seminal capsule being attached to the spermathecal plate (the transparent capsule shown in Figure 1B is a digested and empty one: Yoshizawa et al., 2014). Therefore, if there is only one slot for insemination, as in the spermatheca of Lepinotus, females cannot receive another capsule while digesting one. With the switching valve, the female Neotrogla (and possibly Sensitibilla and Afrotrogla) can selectively use one of two slots for insemination, with leaving the other slot empty. This enables the females to immediately receive an additional seminal package from the same or other males by using the empty slot. Males are predicted to prudently allocate limited resources, such as nutritious seminal gifts and sperm, to multiple females, especially when operational sex ratio is biased to females, rendering female-female competition for male-derived nuptial gifts (i.e, propensity for multiple mating) more intense (Abe and Kamimura, 2015). The male-holding organ (spiny female penis) and female-induced long copulation durations of 40 – 70 hr (compared to just ~50 min for formation of one seminal capsule in close relatives) in Neotrogla (Yoshizawa et al., 2014) thus can be considered as exaggerated adaptations for such escalated competition for nuptial gifts in this group of insects inhabiting highly oligotrophic cave habitats (Lienhard and Ferreira, 2013; Lienhard and Ferreira, 2015; Yoshizawa et al., 2014).

The condition of having two freshly deposited spermatophores at once is comparable to that in multiple sperm storage organs reported for females of some animal groups, such as dung flies, Drosophila, or tephritid fruit flies (Ward, 1993; Pitnick et al., 1999; Twig and Yuval, 2005). Although theory predicts that having multiple sperm stores can be a powerful mechanism for choosing sperm (Hellriegel and Ward, 1998), evidence is scarce for a gain in fitness by actively selecting for particular sperm from among multiple mates (e.g., Demont et al., 2012; Schäfer et al., 2013). In addition, in the case of Neotrogla and related barklice (Lepinotus patruelis), the content of the seminal capsule is digested quite rapidly if not used for fertilization (Wearing-Wilde, 1995; Yoshizawa et al., 2014). Therefore, the switching valve system reported here likely represents an adaptation for direct benefits (i.e., for obtaining more nutrients) rather than for genetic benefits (i.e., for choosing sperm from high-quality males).

Materials and methods

Three species of Neotrogla were examined. We detected little interspecific variation in the basic mechanism of the spermathecal plate.

A virgin female of Neotrogla brasiliensis (Caboclo Cave, Januária, Minas Gerais, Brazil, 12. iii. 2016: Figures 1DF and 2A–C: voucher ID S8KY03) and a copulating pair of N. truncata (Toca dos Ossos Cave, Ourolândia, Bahia, Brazil, 14. i. 2013: Figure 1EG: voucher ID S8KY69: full shape data provided as Video 1) were used for µCT examination (http://dx.doi.org/10.6084/m9.figshare.6741857). Neotrogla brasiliensis was fixed with FAA solution (formaldehyde-acetic acid-alcohol) and N. truncata was fixed with 80% ethanol. Both samples were then stored in 80% ethanol. Dehydration was conducted in ascending order with 80 – 100% ethanol before drying them at the critical point (EM CPD300, Leica, Wetzlar, Germany) to remove water without serious organ shrinkage. Samples were then scanned using synchrotron microcomputed tomography at the BL47XU (Uesugi and Hoshino, 2012) beamline of the Super Photon ring-8 GeV (SPring-8; Hyogo, Japan) using a stable beam energy of 8 keV in absorption-contrast mode. The tomography system consists of a full-field X-ray microscope with Fresnel zone plate optics (Uesugi et al., 2017). The FOV and effective pixel size are 0.11 mm² and 0.0826 μm², respectively. We used semiautomatic segmentation algorithms based on gray-value differences in the software ITK-SNAP (Yushkevich et al., 2006) to obtain 3D representations of the genitalia of Neotrogla. Rendering of the mesh objects was carried out using the software BLENDER (blender.org). Objects were imported as stl files, surface meshes were slightly smoothed, and the number of vertices were reduced without altering the 3D geometry. No further processing was applied. All measurements were carried out in BLENDER.

A virgin female of N. brasiliensis (Figures 1C and 2D: voucher ID: CLKY1) was also used for confocal laser scanning microscope (CLSM) observation (Leica TCS SP5, Wetzlar, Germany). The spermathecal plate was removed and mounted on a glass slide with glycerol. We used an excitation wavelength of 488 nm and an emission wavelength of 510 – 680 nm, detected using two channels and visualized separately with two pseudocolors (510 – 580 nm = green; 580 – 680 nm = red). To visualize resilin, we used an excitation wavelength of 405 nm and an emission wavelength of 420 – 480 nm, detected on one channel and represented with a blue pseudocolor.

A virgin female of N. aurora (Gruta Couve-Flor cave; Aurora do Tocantins, Tocantins, Brazil, 7. i. 2009) was used to take the whole-abdomen photo shown in Figure 1A. The abdomen was removed from a fixed specimen and soaked in Proteinase K at 45°C overnight and stored in glycerol. Photographs were taken with an Olympus E-M5 digital camera attached to an Olympus SZX16 binocular microscope (Tokyo, Japan). Partially focused pictures were combined using Helicon Focus (Helicon Soft Ltd., http://www.heliconsoft.com) to obtain images with a high depth of field. The holotype female of N. truncata (Lienhard et al., 1893) was used for photographing the spermathecal plate shown in Figure 1B. Photographs were taken with an Olympus E-M5 attached to a Zeiss Axiophot compound light microscope (Oberkochen, Germany).

In the interests of transparency, eLife includes the editorial decision letter and accompanying author responses. A lightly edited version of the letter sent to the authors after peer review is shown, indicating the most substantive concerns; minor comments are not usually included.

Thank you for submitting your article "A biological switching valve evolved in the female of a sex-role reversed cave insect to receive more seminal gifts" for consideration by eLife. Your article has been reviewed by Ian Baldwin as the Senior Editor, a Reviewing Editor, and three reviewers. The following individuals involved in review of your submission have agreed to reveal their identity: Liuqi Gu (Reviewer #3).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

The three reviews were all in general agreement that the work provided valuable insights into the valve, but that the work required more careful discussion and development of the biological context regarding a number of key points, which are articulated in the reviews below. The reviews are provided in full below and we would like to see a point-by-point description of how you have addressed the concerns when you submit the revision. While the vast majority of the reviews can be addressed without new data, the last major point of the first review may require new data to address fully.

Reviewer #1:

Authors have detailed the mechanism for sperm storage in Neotrogla. The work they have done is of high quality and given the unique nature of the female intermittent organ in this species, is interesting to biologists interested in sexual selection. The valve described, although relatively simple, does not seem to be used in engineered products (though I am definitely not an expert here), and thus could serve as inspiration for the design of new valves. Although the work here is sound, I do have some concerns, detailed below.

My major issue with the paper is in the explanation, which is light, and interpretation which sometimes over extrapolates from what the authors have found. In particular, (as stated in the Title) the author claim that the switching valve has evolved to receive more seminal gifts. I am not sure I understand the logic here. I think the fact the female has two capsules allows for more sperm storage, I don't think having a switching valve per se means a female can receive more sperm (i.e. if there was no valve, the sperm capsules would fill equally from two matings). Given the clear complexity of the valve described, evolving a multiple / a larger spermatheca seems a much simpler solution to evolve if the only factor driving its evolution is to receive more sperm. This is a key point as the authors interpret their findings in this light, while dismissing other explanations (e.g. Discussion section), and so should be considered more carefully.

The authors also state that females also have 'detached' capsules. Is the sperm in these capsules used for fertilization? Or is only sperm in the attached capsules used for fertilization? This would have consequences for the interpretation of their evolution. Perhaps an advantage of the switching valve would be, that following detachment of a capsule from the spermathecal plate, there is lag time in producing a new capsule (?), and thus by having two there will be more chance of having a capsule ready if they encounter a male soon after one has been detached?

In addition, the paper suggests that a single seminal capsule used in a single mating, however I don't think the authors actually show this (i.e. could be possible that a female switches which sperm capsule she is filling during the long mating period?).

Reviewer #2:

This is a very interesting study on the fascinating booklice genus Neotrogla. These tiny insects are adapted to extreme cave habitats and have evolved exceptional genital structures associated with a sex-role reversal. Males produce large, nutritious spermatophores that the females store in their spermathecae for fertilisation as well as for nutrition. Sperm are apparently taken up by females, which possess a penis like intromitting structure that also secures a firm attachment of the male to the female. Copulations can take several days and the general sexual biology suggests that females compete for nutritious seminal capsules and coerce males into lengthy copulations and perhaps into producing additional sperm packages. Little is known about the ecology and behaviour of the species so that the costs of producing seminal capsules as well as the potential benefits for females to receive multiple capsules must remain speculative.

The current study is concerned with a special valve that females can use to direct the sperm flow. The morphology of female genitals is investigated using μCT and 3D reconstruction as well as confocal laser scanning microscopy.

I have no major concerns and only editorial suggestions. However, not being a morphologist, I had some trouble understanding what happens during copulation. In a previous study, I found highly relevant additional information that was indeed crucial for my comprehension of the highly peculiar sex in this animal group. To the benefit of a non-specialised readership I would therefore advise to add more detail and less technical language about the copulation procedure.

I appreciate that word counts are limited but perhaps some of the information could be included into the figure legends if there is no space in the text. To my taste, parts about valve technology could be replaced to some extend with some more details on sexual biology.

I am fascinated by these animals and am looking forward to additional future insights into their reproductive biology.

Reviewer #3:

The authors present an interesting story about a biological valve in female reproductive system of a cave insect. The manuscript is well composed with clear organization and good writing. The description is on point and the figures are of decent quality. Therefore, I consider its publication in eLife to be appropriate, with no additional work required.

Although, I wish the authors may address the following concern:

Three species of Neotrogla were examined in this study, however, this is not mentioned except in the Materials and methods section, which follows after Discussion section. I think not only it's necessary to describe this information in the main text (Abstract, Introduction, Results section and Discussion section), but also the authors should describe any nuances in morphological difference among the three species examined, if any.

The three reviews were all in general agreement that the work provided valuable insights into the valve, but that the work required more careful discussion and development of the biological context regarding a number of key points, which are articulated in the reviews below. The reviews are provided in full below and we would like to see a point-by-point description of how you have addressed the concerns when you submit the revision. While the vast majority of the reviews can be addressed without new data, the last major point of the first review may require new data to address fully.

Reviewer #1:

Authors have detailed the mechanism for sperm storage in Neotrogla. The work they have done is of high quality and given the unique nature of the female intermittent organ in this species, is interesting to biologists interested in sexual selection. The valve described, although relatively simple, does not seem to be used in engineered products (though I am defiantly not an expert here), and thus could serve as inspiration for the design of new valves. Although the work here is sound, I do have some concerns, detailed below.

Thank you very much for your interests and constructive comments to our paper.

My major issue with the paper is in the explanation, which is light, and interpretation which sometimes over extrapolates from what the authors have found.

We provide now more detailed explanations and toned down some over extrapolations. All of the following major concerns are rooted from our poor explanations for the formation of the seminal capsule. Therefore, our replies are given at the end of the three comments given from reviewer 1. Please also refer to the new main text of the manuscript for more details.

In particular, (as stated in the Title) the author claim that the switching valve has evolved to receive more seminal gifts. I am not sure I understand the logic here. I think the fact the female has two capsules allows for more sperm storage, I don't think having a switching valve per se means a female can receive more sperm (i.e. if there was no valve, the sperm capsules would fill equally from two matings). Given the clear complexity of the valve described, evolving a multiple / a larger spermatheca seems a much simpler solution to evolve if the only factor driving its evolution is to receive more sperm. This is a key point as the authors interpret their findings in this light, while dismissing other explanations (e.g. Discussion section), and so should be considered more carefully.

The authors also state that females also have 'detached' capsules. Is the sperm in these capsules used for fertilization? Or is only sperm in the attached capsules used for fertilization? This would have consequences for the interpretation of their evolution. Perhaps an advantage of the switching valve would be, that following detachment of a capsule from the spermathecal plate, there is lag time in producing a new capsule (?), and thus by having two there will be more chance of having a capsule ready if they encounter a male soon after one has been detached?

In addition, the paper suggests that a single seminal capsule used in a single mating, however I don't think the authors actually show this (i.e. could be possible that a female switches which sperm capsule she is filling during the long mating period?).

All of the above major concerns raised by reviewer 1 are caused by our poor explanation about how the seminal capsule is formed. Our and previous observations showed that capsules are formed within the females’ sperm storage system during copulation. Those observations show that the formation of the capsule is initiated by seminal transfer. Empty capsules are not prepared by the females as the reviewer might have assumed (again this is due to our previous poor explanations which have now been modified). The capsule formation process is least understood (see below), but probably initiated either by a substance provided by the male or by a protective response of the female against the semen (known as “melanisation”). Alternatively, capsule formation could be a result of a chemical reaction between the semen and spermathecal fluid. Because the female spermatheca does not have any muscles or structures enabling sucking up semen, the seminal transfer is apparently caused by an active infusion of the male, as is the case for the majority of animals including insects. When the semen was not used for fertilization, the content of the seminal capsule was consumed by females while the capsule was attached to the spermathecal plate. After consumption of the content of a capsule, the empty capsule detached from the plate before or during the next copulation. The detached capsule is kept within the spermatheca and is never dissolved.

The above knowledge on the seminal capsule formation was added in the revised manuscript (Introduction and Discussion section), which we hope remove the readers' confusion about the formation of the seminal capsule and helps to understand the function of the switching valve. In contrast, the detailed process of the formation of the sperm capsule is less well understood. One possible explanation, although not mentioned previously, is that the formation of the capsule shell can be the result of a chemical reaction between the contents of the semen and the spermathecal fluid (like a reaction between a drop of sodium alginate solution put into calcium chloride solution, which provides solid outer shell containing liquid sodium alginate solution). The size of all the sperm capsules is fairly constant possibly because it is: (1) limited by the amount of semen males can produce, or (2) constrained by the balance between the speed of seminal flow and speed of chemical reaction for formation of the capsule shell. In either case (or combination of both), having two openings WITHOUT a switching valve would cause formation of two half-sized seminal capsules which would occupy both plate slots. This would in turn not increase the female's benefit since multiple matings are not possible while previous sperm capsules are consumed or dissolved. We think this explanation, although likely, is too speculative to discuss in the paper.

Having multiple spermathecae can also be an alternative option to obtain more semen. As also mentioned by reviewer 2 and in our paper, this option is employed by some insects and spiders. In these animals, liquid semen directly fills the spermatheca. In contrast, the evolutionary origin of a hard seminal capsule apparently preceded the evolution of the switching valve in Sensitibillini (formation of the sperm capsule is common in the outgroups). Such differences might have been a factor for the evolution of the switching valve, but it is also too speculative to discuss in the paper.

The main purpose of our paper is to report and describe the biological switching valve and discuss its potential relationship for the evolution of the female penis. Ultimate factors for the origin of the switching valve deserve attention by future investigations and it would be premature to discuss in the present paper. We therefore also modified the title to tone down the over extrapolations about the ultimate factor for the origin of the switching valve.

Reviewer #2:

This is a very interesting study on the fascinating booklice genus Neotrogla. These tiny insects are adapted to extreme cave habitats and have evolved exceptional genital structures associated with a sex-role reversal. Males produce large, nutritious spermatophores that the females store in their spermathecae for fertilisation as well as for nutrition. Sperm are apparently taken up by females, which possess a penis like intromitting structure that also secures a firm attachment of the male to the female. Copulations can take several days and the general sexual biology suggests that females compete for nutritious seminal capsules and coerce males into lengthy copulations and perhaps into producing additional sperm packages. Little is known about the ecology and behaviour of the species so that the costs of producing seminal capsules as well as the potential benefits for females to receive multiple capsules must remain speculative.

The current study is concerned with a special valve that females can use to direct the sperm flow. The morphology of female genitals is investigated using μCT and 3D reconstruction as well as confocal laser scanning microscopy.

Thank you very much for your very positive reaction and useful comments to improve the manuscript. Some of the concerns raised by reviewer 2 are in common with those by reviewer 1. Please also see the reply to reviewer 1, in which detailed explanation about the formation process of the sperm capsule is given.

I have no major concerns and only editorial suggestions. However, not being a morphologist, I had some trouble understanding what happens during copulation. In a previous study, I found highly relevant additional information that was indeed crucial for my comprehension of the highly peculiar sex in this animal group. To the benefit of a non-specialised readership I would therefore advise to add more detail and less technical language about the copulation procedure.

I appreciate that word counts are limited but perhaps some of the information could be included into the figure legends if there is no space in the text. To my taste, parts about valve technology could be replaced to some extend with some more details on sexual biology.

I am fascinated by these animals and am looking forward to additional future insights into their reproductive biology.

We now provided more detailed explanation about the mating biology and sex-role reversals in Neotrogla in the Introduction and Discussion section.

Reviewer #3:

The authors present an interesting story about a biological valve in female reproductive system of a cave insect. The manuscript is well composed with clear organization and good writing. The description is on point and the figures are of decent quality. Therefore, I consider its publication in eLife to be appropriate, with no additional work required.

Many thanks for your interests and positive reaction.

Although, I wish the authors may address the following concern:

Three species of Neotrogla were examined in this study, however, this is not mentioned except in the Materials and methods section, which follows after Discussion section. I think not only it's necessary to describe this information in the main text (Abstract, Introduction, Results section and Discussion section), but also the authors should describe any nuances in morphological difference among the three species examined, if any.

At the end of the Introduction, we added the information that the present study is based on the examination of three species. No notable morphological differences among these species can be detected in our examination, which was also noted at the beginning of the Material and methods section.