Reproductive organs develop sequentially following scaling rules.

(A) Xenacoelomorpha is an early-branching bilaterian lineage of aquatic worms. Phylogeny modified from (Srivastava, 2022). Animal icons are from phylopic.org, and are in the public domain; the dashed line reflects uncertainty in the consensus phylogeny (Cannon et al., 2016; Kapli and Telford, 2020; Philippe et al., 2011). (B) Dorsal view of an adult Hofstenia miamia. (C) Ventral view of an adult Hofstenia miamia; most reproductive structures are visible in this view. (D) Schematized view of the ventral surface of a worm with known reproductive structures illustrated. (E) Timecourse of a representative worm through development, from hatchling to reproductively mature adult. (F) Schematic of timecourse shown in (E) with key reproductive developments illustrated. The first appearance of each organ is highlighted in red. (G) The length of worms increases over time (R2 = 0.91, p < 0.0001), and (H) worms grow proportionally: their length scales with their width (R2 = 0.85, p < 0.0001). Error band shows 95% confidence interval. (I-K) The length of each reproductive organ scales with increases in body size (penis: R2 = 0.70, p < 0.0001; seminal vesicle: R2 = 0.63, p < 0.0001; ovaries: R2 = 0.84, p < 0.0001). Error band shows 95% confidence interval, with zero values excluded from these regressions. (L) Worms with delayed feeding increases had significant delays in the appearance of their seminal vesicle and ovaries, but not the penis (Welch’s t test for date of appearance for penis: p = 0.08, seminal vesicle: p = 0.04, ovary: p < 0.0001; n ≥ 15). (M) Worms with delayed feeding increases had a smaller body length when a penis and seminal vesicle appeared, but not when ovaries appeared (Welch’s t test for length on date of appearance for penis: p = 0.005, seminal vesicle: p = 0.03, ovary: p = 0.74; n ≥ 15). Asterisks indicate statistical significance. (N) Ranking the order in which reproductive organs appear (y-axis) in developing worms reveals a stepwise pattern of reproductive differentiation. The x-axis shows individual worms. dpl = days post laying. Scale bars: 1 mm.

Reproductive organ development follows similar patterns in different growth contexts.

(A) Schematic of regeneration of the penis, seminal vesicle, and ovaries following three different amputations. Shading indicates the tissue that regenerates. (B-J) Growth dynamics of reproductive organs (within column) for each of three amputations (within row). Error bands show SEM. (K) Time course of a starving worm undergoing de-growth and step-wise loss of reproductive organs. (L) Schematic of reproductive organ degradation as seen in (K) over the course of starvation-induced de-growth. (M) Worm length decreases over the course of degrowth (R2 = 0.85, p < 0.0001). Error band shows 95% confidence interval. (N) Worms shrink as they grow; their lengths and widths decrease proportionally (R2 = 0.73, p < 0.0001). Error band shows 95% confidence interval. (O) Across different growth contexts, reproductive organs appear or disappear at roughly consistent body lengths. Ranking the order in which reproductive structures are gained in regenerating worms (Q) and lost in worms undergoing de-growth (P) shows that organs are gained and lost in roughly the same order in all growth contexts. The x-axis shows individual worms in these plots. dps = days post onset of starvation. dpa = days post amputation. Scale bars: 1 mm.

The male reproductive system of Hofstenia includes a penis with a collection of needles, a sperm containing organ (seminal vesicle), and two testes that span the dorsal surface.

(A) A schematized view of the ventral surface of the worm with male reproductive structures highlighted in red. (B) Schematic of male reproductive structures with the copulatory apparatus (excluding the seminal vesicle) highlighted. (C) Labeling with an actin dye (white) labels the male gonopore, sheath, penis stylets, and prostatic vesicle. (D) A histological section also reveals these organs. (E) Schematic of the male reproductive system, with the penis stylets highlighted. (F) The stylets are a bundle of needles labeled by actin. (G) The posterior of the penis sheath terminates in a ring of hair-like projections, also labeled by actin. (H) Schematic of the male copulatory apparatus, with the prostatic vesicle highlighted. (I) Actin staining with a nuclear label (Hoechst) shows that the prostatic vesicle is enveloped by a thin epithelium-like layer, and contains densely packed sperm. (J) Schematic of the male copulatory apparatus, with the seminal vesicle highlighted. (K) The morphology of the copulatory apparatus in mature, adult worms is similar to that of early adults (previous panels). (L) The seminal vesicle of this adult worm contains densely-packed sperm. (M) Dissecting out an adult seminal vesicle allows labeling of individual sperm cells, showing their distinctive morphology. (N) Schematic of a transverse view of an adult worm’s anterior, showing the relative organization of the seminal vesicle and testes. (O) Transverse sections show that testes appear as a continuous structure that spans the dorsal surface of the worm. (P) The testes extend through the dorso-ventral axis of the worm and wrap around the head. The pharynx (labeled and circled with a dotted line) contains residual food. (Q) Schematic of the male copulatory apparatus, with the testes highlighted. (R-T) Nuclear staining on an adult worm, cut sagittally, reveals the testes, which contains dense bundles of sperm organized around clusters of cells in the parenchyma. (U) Histological sections confirm this organization of the testes. Scale bars: 20μm (C, U), 10μm (F-G, M), 50μm (D,I,S, T), 100μm (K,L,R), 200μm (O, P).

Stepwise emergence of individual components of the male reproductive system.

(A) Actin-dye labeling shows how the male reproductive system changes over the course of post-embryonic development (shown here from left to right). The sheath and stylets emerge first, followed by the appearance of the prostatic vesicle. (B) FISH for the male germline marker pa1b3-2 results in two regions of ventrolateral expression that extend along the dorsal-ventral axis to different depths. Images are organized by pseudo-time: from least-developed (and smallest) on the left, to most-developed (and largest) on the right. Panels in (C) show depth-coloration, showing that the testes extend through the dorso-ventral axis. (D) Cross-sections of worms at different points in reproductive development stained with nuclear dye show that testes grow towards the dorsal surface and eventually meet to form one continuous structure. (E) Cartoon schematic of cross-sections shown in (D). Scale bars: 20μm (A), 100μm (B, C), 200μm (D). Estimated worm lengths (wl) are noted under each panel.

The female reproductive system includes embryos stored at the base of the pharynx and oocytes in ovaries.

(A) A schematized view of the ventral surface of the worm with female reproductive structures highlighted. (B) Eggs near the pharynx of the worm (within the red circle) are fertilized and mature while oocytes in ovaries (red arrow) are immature or unfertilized, with a visible germinal vesicle. (C) A sagittal histological section shows that the ovaries contain oocytes of varied size and maturity embedded in the parenchyma. (D) FISH shows that cgnl1-2 labels immature oocytes in the ovaries. (E) Oocytes in ovaries are also labeled by a Piwi-1 antibody. (F) A histological transverse section of an immature oocyte encircled by follicular cells. Inset: sperm cells appear to be trapped in the follicle. (G) Piwi-1 immunofluorescence confirms the organization of follicular cells, and nuclear staining sometimes identifies sperm apparently trapped in its surface (inset). Histology also shows that immature oocytes may have irregular shapes (H), contain a germinal vesicle (H,I), and possess an abundance of (likely yolk) granules (I,J). Blue arrows label germinal vesicles in all relevant panels; yellow arrows label sperm; white arrows label follicular cells. Scale bars: 100μm (B, C (inset), D-F), 500μm (C), 50μm (G, H-J)

H. miamia lays eggs through the mouth and exhibits environmental preferences in egg laying.

(A) Sequence of images from a video of egg-laying through the mouth. Eggs in the pharynx and emerging through the mouth are shaded blue. (B) Schematic showing presumed process of embryo traveling from the cavity beneath the pharynx to the pharynx and then out through the mouth. (C) Histogram showing the timing of eggs laid by adult worms living in communal tanks and then isolated. (D) Histogram showing the timing of eggs laid by worms that undergo reproductive development in isolation and then self-fertilize. (E) Histogram showing the timing of eggs laid by worms that are allowed to mate once. (F) Scatterplot of the percentage of eggs found on the floor of communal tanks (n = 30). This is significantly different from the expected percentage of eggs based on tank surface area (t-test p < 0.0001). (G) Kernel density estimate of egg locations on a subset of tank surfaces with similar dimensions (n = 2144 eggs). (H) Density-based spatial clustering of egg coordinates shows that eggs are laid in clutches. Number of eggs in a clutch shown in white. (I) In some culturing conditions, worms lay clutches of up to 145 eggs. (J) New worms add eggs to pre-existing clutches laid by other worms. (K) Worms that are unfed for 4 days lay fewer eggs than fed worms (n = 9 tanks, t-test p < 0.0001). (L) Unfed worms that are subsequently fed lay more eggs than worms that are continuously fed (n ≥ 8 tanks, t-test p < 0.0001).

Reproductive life histories in Acoelomorpha.

(A) The life cycle of Hofstenia miamia, with major reproductive events displayed. (B) Family-level phylogeny of Acoelomorpha (Nemertodermatida, Acoela) showing anatomical and reproductive life history traits (Table S1): position of the mouth, whether gonads are mixed or separated by sex, penis type, paired or unpaired testes, paired or unpaired ovaries, presence or absence of a female gonopore, presence or absence of a seminal bursa, the number of associated bursal nozzles, egg-laying mode, mode of sexual reproduction, alternative reproductive strategies, and regenerative capacity (see Table S4 for definitions of terms and categories). Schematic diagram of the reproductive anatomy of representative species from each family within Acoelomorpha with specific structures colored: male copulatory organ (purple), sperm in testes and/or seminal vesicle (blue), oocytes (red), female or shared gonopore and/or bursa (green). (Table S1). White boxes represent unknown phenotypic states, and in the case of asexual reproduction, its possible absence. Phenotypic classifications are from (Achatz and Hooge, 2006; Apelt, 1969; Bailly et al., 2014; Beltagi and Mandura, 1991; Bock, 1923; Boone et al., 2011; Bush, 1975; Costello and Costello, 1939, 1938; Crezee, 1978; Dörjes, 1968, 1966; Faubel, 1976, 1974; Faubel and Cameron, 2001; Gardiner, 1895; Grae and Kozloff, 1999; Hooge, 2003; Hooge et al., 2007; Hooge and Smith, 2004; Hyman, 1937; Kostenko, 1989; Kozloff, 2000b; Meyer-Wachsmuth et al., 2014; Peebles, 1915; Perea-Atienza et al., 2013; Raikova et al., 1995; Riser, 1987; Shannon and Achatz, 2007; Steinböck, 1966; Sterrer, 1998; Watzin, 1984).

Body length is a better predictor of organ length than age in multiple regression.

Multiple regression coefficients and associated p-values are reported for each organ. Wald’s test was used on coefficients to test whether they are significantly different from each other.

ANCOVA of the relationship between organ size and body size.

Glossary of anatomical and reproductive traits.

Gross reproductive morphology is visible from the ventral surface of H. miamia.

(A) The male reproductive system includes the male gonopore, penis, prostatic vesicle, and seminal vesicle. (B) H. miamia’s ovaries are lateral, and organized along the anterior-posterior axis. There is no visible organization of oocytes by size or stage within the ovaries. (C) A cluster of zygotes is visible medially, and immediately posterior to the male copulatory apparatus. (D) Representative time course of embryonic development of embryos dissected from the central cavity. Scale bars: 1 mm.

Growth and scaling.

(A) Time course of feeding changes through juvenile development, showing the delayed increases in feeding for one group, and the introduction of brine shrimp late in development. (B) Survival curve showing the number of worms in each of the two feeding treatments over time. (C) The aspect ratio (length:width) of worms is consistent over time (slope = 0.002, R2 = 0.002, p = 0.31), showing that worms grow proportionally over development. Error band shows 95% confidence interval. (D) Worm length increases over time, with worms growing more slowly when their increased feeding was delayed. Error band shows SEM. (E) Residuals of ovary length (regressed against body length) within each worm are highly correlated (Pearson correlation coefficient = 0.7, p < 0.0001). Error band shows 95% confidence interval.

Across different growth contexts, body and organ scaling rules are conserved.

(A, C, E) Representative time course of regenerating tail tip, sagittal cut, and head, and (B, D, F) schematized version with reproductive structures shown. (G) Regenerating heads and tail tips increase in length following amputation while sagittally-cut fragments shrink and then begin to grow. Error bar shows SEM. (H) Ovaries scale with body length in sagittally-cut fragments. (I) Ovaries regenerate once a tail fragment reaches a certain length. Scale bars: 1 mm.

Across different growth contexts, body and organ scaling rules are conserved.

(A) Worm length decreases over time when worms are starved (starved: R2 = 0.85, p < 0.0001; fed control: R2 = 0.004, p = 0.48). Error band shows 95% confidence interval. (B-D) Penis length (starved: R2 = 0.54, p < 0.0001; fed control: R2 = 0.04, p = 0.03), seminal vesicle length (starved: R2 = 0.60, p < 0.0001; fed control: R2 = 0.03, p = 0.06), and mean ovary length (starved: R2 = 0.67, p < 0.0001; fed control: R2 = 0.02, p = 0.15) decrease over time for worms experiencing de-growth but do not change in worms that are continuously fed. Error band shows 95% confidence interval. (E) The number of zygotes in the central cavity decreases over time in both starved and fed worms, likely because these worms are isolated (starved: R2 = 0.48, p < 0.0001; fed control: R2 = 0.33, p < 0.0001). Error band shows 95% confidence interval. (F) When controlling for the effect of body size, the residuals of ovary length (regressed against body length) are strongly correlated between ovaries within worms over the course of de-growth (Pearson correlation coefficient = 0.75, p < 0.0001). Error band shows 95% confidence interval. (G-I) The size of reproductive structures scale similarly (see Table S3) across development, de-growth, and regeneration (penis R2 = 0.70, p < 0.0001; seminal vesicle: R2 = 0.50, p < 0.0001; ovary: R2 = 0.76, p < 0.0001). Error band shows 95% confidence interval. (J-L) The residuals of organ length (regressed against body length) generally converge towards zero over the course of regeneration despite sometimes displaying large initial deviations. (M) Sagittal fragments degrade their existing ovary during initial regeneration at a faster rate normalized to body size (slope = -0.015) than adult worms experiencing de-growth (slope = -0.003) (Wald’s test on slope of linear regression of ovary length and day: p=0.0006). Error band shows 95% confidence interval.

Fine structure of H. miamia’s male reproductive system.

(A) The gross morphology and location of the reproductive system can be seen in sagittally-cut worms with nuclear staining. (B-C) Penis stylets are arranged in a bundle, seen with differential interference contrast (DIC). (D) Schematic of the male reproductive structures with the copulatory apparatus highlighted. (E) Antibody staining for FMRFamide shows that the penis sheath is surrounded by presumptive neurons. (F) Immunofluorescence for tropomyosin, and actin and nuclear labeling enables visualization of the musculature surrounding the male structures. (G) A packet of sperm ejected during mating made up of (H) sperm cells. (I) Schematic of the male reproductive structures with the testes highlighted. (J) Sperm in the testes develop in clusters interspersed with the muscle of the body, visualized by immunofluorescence for tropomyosin, a muscle marker. Scale bars: 500μm (A), 100μm (B, G), 50μm (C, E, J), 25μm (F), 10μm (H).

Histology of the male reproductive system reveals the organization of the male reproductive system along different body axes.

(A) The male reproductive system in a sagittal section of a juvenile worm has a male gonopore but lacks a clear male copulatory apparatus, prostatic vesicle, and seminal vesicle. (B) A longitudinal section showing the male gonopore. (C-E) Sagittal sections through the male reproductive system along the medial-lateral axis showing a channel of sperm (yellow arrows) connecting the prostatic vesicle and seminal vesicle. (F-H) Transverse sections moving along the anterior-posterior axis show the positions of the prostatic vesicle and seminal vesicle and the musculature and glands surrounding the seminal vesicle. Scale bars: 100 um.

Testes emerge as lateralized regions and grow to span the dorsal surface as worms grow.

(A) FISH for the male germline marker pa1b3-2 results in two regions of ventrolateral expression that extend toward the dorsal surfaces. (B) Hoechst and pa1b3-2 mRNA expression label sperm cells. (C) Nuclear staining of cross-sections at different points along the anterior-posterior axis within the head shows that the testes emerge as two lateral lobes that gradually form a continuous cylindrical structure. Cartoons depict the plane of sectioning (red line) and reproductive structures (white). Scale bars: 250 μm (A), 50μm (B), 200μm (C).

Oocytes in Hofstenia’s ovaries are surrounded by follicular cells regardless of position or maturity.

(A) Nuclear staining on a sagittal-cut worm confirms the presence of ovaries along the ventral side of the worm. Consistent with our other imaging, ovaries lack any sense of organization in regards to maturity along the anterior-posterior axis. (B-D) Oocytes are surrounded by a layer of follicular cells. Scale bars: 500μm (A), 100μm (D).

Worms have spatial preferences during egg-laying that are robust to suboptimal conditions.

(A) Quantifying the time from the start of visible muscle contractions to egg laying reveals that the egg-laying process spans a timescale of seconds to minutes for each egg, with most eggs being laid within 3 minutes. (B) Kernel density estimate of egg locations on the short walls of the worm tanks (n = 1390 eggs). (C) Random observations of worms in their culture tanks (n = 12 tanks) finds only a minority of worms on the upper half of tank walls, even in conditions where we expect them to be actively laying eggs. (D) Worms preferentially avoid laying eggs on the floor of the tank regardless of whether they are fed or unfed (two-sample t-test: p= 0.62, n ≥ 8). (E) Distribution of clutch sizes in communal tanks (n = 42 clutches) shows that clutches frequently contain more than 10 eggs laid within a 3-4 day window. (F) Distribution of eggs laid by individual worms across different egg laying contexts shows that most worms lay fewer than 10 eggs in a 3-4 day window (G) Worms often add eggs to clutches constructed by other worms; the distribution shows the numbers of new eggs added to old clutches.