Comparative morphological analysis of compound eye miniaturization in minute hymenoptera

Accepted Manuscript Comparative morphological analysis of compound eye miniaturization in minute Hymenoptera Anastasia Makarova , Alexey Polilov , Stefan Fischer PII:

S1467-8039(14)00102-9

DOI:

10.1016/j.asd.2014.11.001

Reference:

ASD 598

To appear in:

Arthropod Structure and Development

Received Date: 16 April 2014 Revised Date:

31 October 2014

Accepted Date: 5 November 2014

Please cite this article as: Makarova, A., Polilov, A., Fischer, S., Comparative morphological analysis of compound eye miniaturization in minute Hymenoptera, Arthropod Structure and Development (2014), doi: 10.1016/j.asd.2014.11.001. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Comparative morphological analysis of compound eye miniaturization in minute Hymenoptera ANASTASIA MAKAROVA1, ALEXEY POLILOV1 AND STEFAN FISCHER2 1

Department of Entomology, Faculty of Biology, Lomonosov Moscow State University, Moscow, Russia

2

Department of Psychology and Neuroscience, Life Sciences Centre, Dalhousie University, Halifax, Nova

RI PT

Scotia, Canada

Keywords: Megaphragma, Trichogramma, Anaphes, insects, vision.

SC

Abstract

Due to their small size, diminutive parasitic wasps are outstanding subjects for

M AN U

investigating aspects of body miniaturization. Information on minute compound eyes is still scarce, and we therefore investigated eye morphology in one of the smallest known Hymenopteran species Megaphragma mymaripenne (body size 0.2 mm) relative to Anaphes flavipes (body size 0.45 mm) and compared the data with available information for Trichogramma evanescens (body size 0.4 mm). The eyes of all three species are of the apposition kind, and each ommatidium possesses the typical cellular of

ommatidia

found

TE D

organization

in

larger

hymenopterans.

Compound

eye

miniaturization does not therefore involve a reduction in cell numbers or elimination of cell types. Six size-related adaptations were detected in the smallest eyes investigated, namely a) a decrease in the radius of curvature of the cornea compared with larger

EP

hymenopterans; b) the lack of extensions to the basal matrix from secondary pigment cells; c) the interlocking arrangement of the retinula cell nuclei in neighboring

AC C

ommatidia; d) the distal positions of retinula cell nuclei in M. mymaripenne; e) the elongated shape of retinula cell pigment granules of both studied species; and f) an increase in rhabdom diameter in M. mymaripenne compared with A. flavipes and T. evanescens. The adaptations are discussed with respect to compound eye miniaturizations as well as their functional consequences based on optical calculations.

1. Introduction Structural peculiarities related to miniaturization can be traced in anatomy of many small insects (Polilov, 2015). The extremely small size affects most of the organ systems of 1

ACCEPTED MANUSCRIPT

tiny hymenoptera: the skeletal and the muscular system, as well as the reproductive (Polilov, 2007) and the nervous system (Makarova, Polilov, 2013). In the smallest insects modifications occur not only at the level of organs, but also at the cellular level (Polilov, 2012). Initial investigations on structural and functional changes in miniature compound eyes started about 10 years ago on eyes of small scarabaeid beetles (Meyer-Rochow and Gal, 2004) while more recent work dealing with lepidopteran

RI PT

compound eyes revealed morphologically intermediate eyes, in between apposition and superposition types (Honkanen and Meyer-Rochow, 2009; Fischer et al., 2011, 2012a,b, 2013).With respect to vision in the tiniest hymenopterans, to date only one study has been published on the parasitoid wasp Trichogramma evanescens

SC

(Westwood 1833) with a body size of only 0.3-0.4 mm (Fischer et al., 2011). Work on the optics of this tiny species revealed size-related adaptational changes in eye morphology, and especially the very precise, alternating positions of the nuclei of

M AN U

regular retinula cells suggests that the described eye comes close to a fundamental limit with respect to the available space within the eye and its functional design (Fischer et al., 2011).

Nevertheless there exist species even smaller in size than Trichogramma, like Megaphragma mymaripenne (Timberlake 1924), with a body size around 0.2 mm. The

TE D

question therefore arises how are organized and if special adaptations or even reductions in cell types are to be found. In order to study not only the minimal possible limits, but also questions regarding the priority with which different parameters have had

EP

an impact on eye design, a comparative investigation was initiated that included different species of parasitoid wasps of small body- and eye-sizes. We investigated the eye morphology of one of the smallest known Hymenopteran species Megaphragma

AC C

mymaripenne (body size 0.2 mm) and of a larger species Anaphes flavipes (body size 0.45 mm) and compared these with available information for Trichogramma evanescens (Fischer et al., 2011).

2. Material and methods 2.1. List of taxa examined. Adult females of two small parasitoid wasp species were studied. Specimens of the mymarid Anaphes flavipes (Foerster 1841) (Mymaridae) were collected in the Moscow region (2009) and specimens of Megaphragma mymaripenne (Timberlake 1924) (Trichogrammatidae) in Blanes, Spain (2008) and 2

ACCEPTED MANUSCRIPT

Funchal, Madeira, Portugal (2009). M. mymaripenne were hatched from eggs of Heliothrips haemorrhoidalis (Bouché 1833), which were collected in July 2008 in Blanes and October 2009 in Madeira on Viburnum tinus L., 1753 (Adoxaceae). A. flavipes were collected with an insect net in places of possible habitat. 2.2. Light microscopy (LM). Light-adapted specimens were fixed in formaldehyde–

RI PT

ethanol–acetic acid (FAE) and embedded in Araldite M. The samples were then sectioned at 1 µm with a Leica microtome (RM 2255) and stained with toluidine blue. The serial sections were photographed with a Tucsen digital camera on an Olympus

SC

BX43.

2.3. Transmission electron microscopy (TEM). The sample material was fixed in 2% glutaraldehyde solution in 0.1 M phosphate buffer (pH 7.2) for 12 hours and post-fixed

M AN U

with 1% osmium tetroxide for 2 hours in the same buffer. All specimens were fixed in the light-adapted state. The specimens, en-bloc stained with uranyl-acetate, were embedded in Epon 812 and sectioned with a Leica UC6 ultramicrotome at a thickness of 50 nm. After ultrathin sections were stained with lead citrate for 10-12 minutes, the samples were investigated with a Jeol JEM-1011 transmission electron microscope

TE D

operated at 80 kV.

2.4. Scanning electron microscopy (SEM). After ultrasonic cleaning, the specimens were critical point dried (Hitachi HCP-1) and coated with a layer of gold in a sputter

EP

coater (Hitachi IB-3). The samples were observed using a Jeol JSM- 6380 SEM,

AC C

operated at 20 kV.

2.5. Measurements. All measurements were made using software (Image J, Rasband, W.S, U.S. National Institutes of Health, Bethesda, MD). Eyes, 3-5 for each species, were used for measurements from the SEMs. For examinations involving LM and TEM, 6-10 eyes of either species were used. SEM was used to determine the dorso-ventral and antero-posterior extent of the eye, the total number of ommatidia per eye, their facet diameters, and the diameters of the ocelli. Longitudinal ultrathin sections for TEM were used to measure ommatidial lengths, the radii of curvature of eye and corneal facets, cone lengths, corneal thicknesses and to investigate the shapes and positions of the pigment granules within the retinula cells, as well as determine the interommatidial angles. Rhabdom diameters, shapes, and diameters of pigment granules of primary 3

ACCEPTED MANUSCRIPT

pigment cells (PPC) and secondary pigment cells (SPC) as well as retinula cells were measured from TEM cross sections. Measurements of rhabdom diameters were taken at the distal tip of the rhabdom. 2.6. Optical calculations

were used to calculate relevant parameters.

RI PT

To compare the optical properties of the compound eyes, the anatomical measurements

The focal lengths of the lenses were calculated on the assumption that the refractive indices in the crystalline cones and the cornea were homogenously distributed, using

l

1

2

3

with

P1 =

n − n , = n′ − n , P r r P l

l

2

2

=−

t PP n 1

2

l

TE D

1

3

M AN U

P =P +P +P

SC

the thick lens formula (Jenkins and White, 1976):

The addition of the powers given by the front and back surfaces of the lens (P1, P2) provides the total lens power of the system. The indices n, nl and n’ describe the refractive indices of the air (n=1), lens and image space; t is the thickness of the lens.

EP

The radii of curvature of the front and back surfaces of the lens are defined by r1 and r2, respectively. An investigation of the properties of the crystalline cones of these species

AC C

appeared near to impossible because of their small size and the uncertainties of interference microscopy on such small structures (Kunze, 1979). Approximations were therefore taken from measurements on the compound eye of another hymenopteran, namely Apis mellifera (Varela and Wiitanen, 1970) for the lens (nl =1.452) and for the cone (n´=1.348).

The focal length (f) and the image focal length (f´) can then be calculated by:

f=

n Pl

and

f′=

n′ Pl

4

ACCEPTED MANUSCRIPT

The focal length (f) was calculated from the secondary nodal point N, the position of which can be determined by measuring the distance to the vertex of the back surface of the lens (Stavenga, 2003):

n′(1− tP1 / nl ) −f Pl

RI PT

dn =

The focal length allowed us to calculate the F-number of the compound eye, which serves as a value to compare the optical properties of different compound eyes (e.g.

SC

Warrant and McIntyre, 1993). The F-number is defined as the ratio of the facet diameter

M AN U

(A) to the focal length (A/f).

Furthermore the acceptance angle of the rhabdom ∆ρrh (diameter of the rhabdom / focal length) was calculated as well as the half-width of the Airy Disk (∆ρl = λ/A) with A= facet diameter and λ as the wavelength of light (Snyder, 1977, 1979). A wavelength of 0.5 µm (green light) was used in the calculation.

TE D

In order to determine the sensitivity of the ommatidia, the sensitivity formula was applied, modified for white light by Warrant and Nilsson (1998):

 π  2 2  d   kl  SW =   A     4  f   2.3 + kl 

EP

2

where A=facet diameter, d=rhabdom diameter, f=focal length, k=absorption coefficient

AC C

(0.0067 µm-1 for invertebrates) and l=rhabdom length.

3. Results

The morphology of the compound eyes of the species investigated will be described in detail for Megaphragma mymaripenne and Anaphes flavipes. Measurements are summarized in comparison to available data for females of Trichogramma evanescens from Fischer et al. (2011), see Table 1.

5

ACCEPTED MANUSCRIPT

3.1. Megaphragma mymaripenne (Trichogrammatidae)

The compound eyes of the parasitoid wasp Megaphragma mymaripenne (Timberlake 1924) are oval in shape (Fig. 1A) and measure about 52.6 µm in dorso-ventral extent. Their anterior-posterior extent is on average 30.7 µm. Each eye has 29 facets with an average diameter of each facet of 8.1 µm. The corneal surface of the facets is smooth

RI PT

(Fig. 1B) and areas of flat cuticle that form a hexagonal lattice separate the strongly curved facet lenses (Fig. 1B). Two interfacetal hairs are present in the apical third of each eye and one large hair is located in the distal part, directly above the eye (Fig. 1A).

7.1 µm in diameter. 3.1.1. The fine-structural organization

SC

In addition to the two compound eyes, three ocelli are present (OC) (Fig. 1A), all about

M AN U

Each ommatidium is made up of the distally located dioptric apparatus (a biconvex lens and crystalline cone), surrounded by primary and secondary pigment cells, and the proximally located photoreceptive retinula cells (Fig. 2D). The total length of the ommatidium is 22.7 µm on average (Fig. 2A) and the interommatidial angle between neighboring ommatidia is 21.5°.

TE D

3.1.1.1. Dioptric apparatus and the primary pigment cells (PPC) From the apex of the cornea to the distal tip of the crystalline cone, the dioptric apparatus is 7.4 µm long. The thickness of the cornea reaches its maximum of 2.6 µm

EP

in the center of the facet. The outer radius of curvature of the cornea measures 3.4 µm, whereas the inner radius of curvature is smaller, on average 2.8 µm (Fig. 2B). With 0.3 µm, the epicuticle in the interommatidial spaces is much thinner. Four cone cells form

AC C

the on average 5.3 µm long eucone crystalline cone, which is surrounded by two primary pigment cells (PPC) (Fig. 2B). From its distal diameter of 5.2 µm, the crystalline cone tapers towards its proximal tip to a width similar to that of the rhabdom. Cone cell projections adjoin the rhabdom (Fig. 2E) and terminate in thickened endings at the basal matrix. The nuclei of the PPC are situated in one plane beneath the maximal cone diameter (Fig. 2B, D, H). Beside the nuclei, the cytoplasm of the PPC contains spherical pigment granules (PG) of high electron density with a diameter of about 0.45 µm (Fig. 2E). 3.1.1.2. Secondary pigment cells (SPC)

6

ACCEPTED MANUSCRIPT

Six secondary pigment cells surround each ommatidium. They are situated in spaces between the distal third of the primary pigment cells of neighboring ommatidia and the cornea (Fig. 2B, D, H). The rather small SPCs are confined to this area and do not extend to the basal matrix. Their nuclei show a high degree of compaction of their chromatin and have a diameter of 1.5 µm in their cross-section (Fig. 2B, D, H). Pigment granules, measuring 0.28 µm in diameter, are present in addition to the nuclei in the

3.1.1.3. Retinula cells and rhabdom

RI PT

SPCs (Fig. 2B, H). Other cell organelles could not be detected.

The ommatidia of M. mymaripenne possess a rhabdom of the fused type, formed by

SC

eight retinula cells over the entire length of the rhabdom (called regular retinula cells) (Fig. 2E). A ninth retinula cell participates only in the most proximal part of the

M AN U

ommatidium (Fig. 3B) and is labeled as irregular retinula cell. Extending from the proximal tip of the cone down to the basal matrix, the rhabdom has a length of 13.3 µm. Its maximal diameter is found in the most distal tip, measuring on average 2.4 µm (Fig. 2G). The microvilli forming the rhabdomeres of the rhabdom have an average diameter of 57.4 nm. Longitudinal sections of the ommatidia show that the nuclei of most regular retinula cells are placed in the distal part of the cells. Some nuclei even make use of the

TE D

space in between the primary pigment cells of neighboring ommatidia (Fig. 2D, G). In deeper levels the nuclei of the retinula cells show alternating positions, which allow an interlocking of neighboring retinula cells (Fig. 3C). This can be detected within one ommatidium as well as also between adjacent ommatidia. The retinula cells contain

EP

pigment granules that are elongated in shape, 0.28 µm in diameter and 0.73 µm in length (Fig. 2D, G). The granules are oriented predominantly with their long axis parallel

AC C

to the axis of the ommatidium and are found close to the cisternae of the endoplasmatic reticulum. The cisternae surround the rhabdom as a narrow sheath (Fig. 2E, white arrows). In the distal part of the retinula cells on average 13 ± 5 pigment granules per retinula cell (n=80) can be found close to the cisternae in cross sections. Typical cellular organelles of retinula cells, such as mitochondria, are mostly located near the cell borders to neighboring ommatidia (Fig. 2E, F; 3C). Tracheoles were not detected neither in the eye nor below the basal matrix. 3.2. Anaphes flavipes (Mymaridae) The eyes of the Mymarid Anaphes flavipes (Foerster 1841) are drop-shaped, with the broadest expansion in the dorsal part (Fig. 1C). The eye’s dorso-ventral extent is about 7

ACCEPTED MANUSCRIPT

97.2 µm, and the average anterior-posterior extent 75.9 µm. Each compound eye contains 121 facets with a diameter of about 8.1 µm. The facets are hexagonally shaped at the base, but they show a more spherical appearance due to the strongly convex cornea (Fig. 1D). The surface of the cornea is smooth and small spaces, 0.64 µm wide, are noticeable between the facets (Fig. 1D). These spaces bear about 15 interfacetal hairs, mainly found in the center of the eye. The average length of the

RI PT

ommatidium is about 33.4 µm and the interommatidial angle is 15° (Fig. 4A, G) . In addition to the two compound eyes, three ocelli are present: the circular vertex ocellus has a diameter of 8 µm, and the two lateral ones, elongate in shape, measure about 12.5

(Fig.

3.2.1. The fine-structural organization

1C).

SC

µm

M AN U

3.2.1.1. Dioptric apparatus and primary pigment cells

The multi-layered and biconvex-shaped corneal lenses (Fig. 4A, B, C, G) possess a strongly bulged outer surface with a radius of curvature of 5.9 µm. The inner radius of curvature is about 5.0 µm. In the center of the facet the cornea reaches its maximal thickness of about 3.5 µm, whereas the thickness of the epicuticle in interommatidial

TE D

spaces does not exceed 1 µm (Fig. 4B). Attached to the proximal side of the cornea is the crystalline cone, which is of the eucone type. It is formed by four cone cells (Fig. 4C) and has an average length of 10 µm. In the distal part, the crystalline cone has a diameter of about 7.1 µm and tapers towards the tip of the rhabdom (Fig. 4B). Cone cell

EP

projections adjoin the rhabdom (Fig. 4E, F) down to the basal matrix. The cone cell nuclei are characterized by a high percentage of heterochromatin (Fig. 4B).

AC C

The crystalline cone is surrounded by two primary pigment cells (Fig. 4B, G), bearing numerous pigment granules. The granules are not homogenous in shape and vary from spherical to ellipsoidal. Their respective average diameters and lengths are 0.44 µm and 0.79 µm, respectively. The nuclei of the PPC are situated in a plane below the maximal diameter of the crystalline cone (Fig. 4B). 3.2.1.2. Secondary pigment cells Six secondary pigment cells (SPC) surround each ommatidium. They are situated directly below the cornea and occupy the space between the primary pigment cells of neighboring ommatidia (Fig. 4B, C). Extensions of the SPC to the basal matrix were not 8

ACCEPTED MANUSCRIPT

seen and the most proximal extension of the secondary pigment cells was found in the plane where the nuclei of the PPC are situated. Despite the small size of the SPC, pigment granules are present in addition to the nucleus. The shape of pigment granules varies from spherical to elongate, measuring 0.42 µm in diameter and 0.75 µm in length on average.

RI PT

3.2.1.3. Retinula cells and rhabdom The ommatidia of A. flavipes possess a centrally fused rhabdom (Fig. 4D, E, F), with a total length of 19.8 µm and a diameter of 1.4 µm (measured in the most distal part). The rhabdom is formed by nine retinula cells, of which eight participate in the formation of

SC

the rhabdom over its entire length from the proximal tip of the cone down to the basal matrix (Fig. 4E). The nuclei of neighboring retinula cells (neither those of the same ommatidium nor those of neighboring ommatidia) are usually not found in the same

M AN U

plane (Fig. 4D), except in the most distal part of the retinula cells. The ninth retinula cell contributes to the rhabdom only in the most proximal part of the ommatidium (Fig. 3A). At this level also tracheoles can be found within the ommatidium (Fig. 3A, 4H, white arrow). The cell organelles of the retinula cells are mainly located near the border of the cells to the neighboring ommatidia. The ellipsoidal pigment granules of the retinula cells

TE D

measure about 0.23 µm in diameter and about 0.69 µm in length. They are oriented predominantly parallel to the longitudinal axis of the ommatidium, along the cisternae of the endoplasmatic reticulum. In the distal part of the retinula cells on average 5 ± 2 pigment granules per retinula cell (n=80) surround the cisternae (counted in cross-

EP

sections). The cisternae are wider compared with those found in M. mymaripenne (Fig. 4H, Fig. 2E). The microvilli forming the rhabdom have a diameter of 72.9 nm.

AC C

3.3 Optical properties

In order to compare the optical features of the investigated species, several parameters were calculated (Table 2). The focal length (f) and image focal length (f´) were calculated using the inner and outer radii of curvature of the lens, its maximal thickness as well as refractive indices for the outside (air n=1), lens (nl =1.452) and crystalline cone (n´=1.348), the latter two values taken from Varela and Wiitanen, 1970. For a facet lens of M. mymaripenne with P1: 0.133 µm-1, P2: 0.037µm-1, P3: -0.009 µm-1 and Pl: 0.161 µm-1 we calculate a focal length (f) of 6.2 µm and an image focal length 9

ACCEPTED MANUSCRIPT

(f´) of 8.4 µm. For A. flavipes a focal length of 10.7 µm and an image focal length of 14.4 µm were computed on the basis of P1: 0.077 µm-1, P2 : 0.021 µm-1, P3: -0.004 µm-1 and Pl: 0.094 µm-1. In both species the lens power is mainly determined by the front surface of the lens (P1). Given the average length of the crystalline cones (M. mymaripenne: 5.3 µm and A. flavipes: 10.0 µm) and the distance of the nodal points to the vertex of the back surface of the lens (M. mymaripenne: 0.17 µm and A. flavipes:

RI PT

0.99 µm), the calculated focal lengths (f) place the focal plane close to the distal tip of the rhabdom in both species. Fischer et al. (2011) also report this to be the case for T. evanescens.

Using the calculated focal lengths and the values of the facet diameters, the F-number

SC

resulted in values of 0.8 for M. mymaripenne and 1.3 and 1.4 for A. flavipes and T. evanescens, respectively.

M AN U

Acceptance angles of the rhabdoms ∆ρrh were calculated as 22.2° for M. mymaripenne, 8.1° for T. evanescens and 4.8° for A. flavipes. The values for the half width of the blur circle ∆ρl calculated for green light (500 nm), are 3.5° for M. mymaripenne and A. flavipes and 4.5° for T. evanescens. The sensitivity formula, modified for white light perception by Warrant and Nilsson (1998), revealed a distinctly higher value of 0.23 (0.06 µm2/sr). 4. Discussion

TE D

µm2/sr for M. mymaripenne than those of A. flavipes (0.04 µm2/sr ) and T. evanescens

EP

We examined the ultrastructure of the compound eyes of two tiny hymenopterans, including one of the currently smallest known insects, Megaphragma mymaripenne (Timberlake 1924). The major goal has been to work out size-related differences in the

AC C

compound eye morphology as well as the resulting functional consequences, that can be put in relation to the extreme spatial constraints imposed by these tiny eyes. 4.1. General organization of the compound eyes – external features Eye size (both dorsal-ventral and anterior-posterior) decreases with body size. This was also shown previously for larger specimens of several groups of insects (Yagi and Koyama, 1963; Menzel and Wehner, 1970; Jander and Jander, 2002; Rutowski et al., 2009; Döring and Spaethe, 2009; Fischer et al., 2013). Different to the mentioned studies, the facet number does not show an allometric reduction with body size in the small species investigated. M. mymaripenne has the smallest body size and the fewest 10

ACCEPTED MANUSCRIPT

ommatidia. In the case of A. flavipes (body size 0.45 mm) and T. evanescens (body size 0.3-0.4 mm), on the other hand, the differences are rather small with respect to their eye sizes with means of 121 facets and 128, respectively. The small difference is caused by a larger facet diameter of 8.1 µm found in A. flavipes compared with 6.39 µm in T. evanescens. In addition to the compound eyes, it is also worth mentioning, that M. mymaripenne possesses the smallest ocelli (diameter of 7.1 µm) of all investigated

RI PT

species. 4.2. Anatomical details

Despite their small size and minute compound eyes, all small hymenopterans

SC

investigated so far (this study and Fischer et al., 2011) show the typical cellular organization of ommatidia found in hymenopterans in general (e.g. Varela and Porter, 1969; Perrelet, 1970; Menzel, 1972). The ommatidia consist of two primary pigment

M AN U

cells, six secondary (interommatidial) pigment cells, four cone cells and nine retinula cells. As in the smallest lepidopterans investigated (Fischer et al., 2013), no size-related reductions in cell numbers are found compared with larger species. Miniaturization therefore has to be performed in other ways than in gaining space by reducing cell numbers or cell types.

TE D

4.2.1. Size-related adaptations

The following sections will deal specifically with adaptations found as result of miniaturization in the eyes of the investigated species. Overall, we found six size-related

EP

adaptations, namely a) a decrease in the radius of curvature of the cornea compared with larger hymenopterans; b) the lack of extensions to the basal matrix from secondary

AC C

pigment cells; c) the interlocking arrangement of the retinula cell nuclei in neighboring ommatidia; d) the distal positions of retinula cell nuclei in M. mymaripenne; e) the elongated shapes of the retinula cell pigment granules; and f) the increased rhabdom diameter in M. mymaripenne compared with A. flavipes and T. evanescens. All adaptations will be discussed in turn. 4.2.1.1. Decrease in the radius of curvature of the cornea compared with larger hymenopterans With decreasing size of the facets a decrease in the radius of curvature is recognizable. In contrast to the flat surface of compound eyes of larger hymenopterans (see e.g. Menzel et al., 1991) with their regular patterns of hexagonal-shaped facets, the 11

ACCEPTED MANUSCRIPT

decrease in the outer radius of curvature of the cornea results in more convex facets. Our calculations, using the thick lens formula, showed that the lens power is mainly determined by the front surface of the lens and thus the outer radius of the facet plays a major role in achieving a short focal length. This is critically needed because of the short dioptric apparatus found in the tiny species.

RI PT

The minimal limit of the outer radius of curvature should roughly be set by half the diameter of the facet. In the case of M. mymaripenne the value of the outer radius of curvature is slightly below half of the diameter of the facet. In order to achieve shorter focal lengths therefore a further decrease in the radius of curvature should only be possible by a decrease in facet diameter. This however has a functional impact,

SC

because diffraction in general increases with smaller facet diameters (Barlow, 1952) and it also decreases the interommatidial angle and thus also sensitivity of the

M AN U

ommatidium. The investigated eyes, however, are not diffraction limited, as the width of the blur circle falls below the acceptance angle of the rhabdom (due to the short focal length and the diameter of the rhabdom).

4.2.1.2. Lack of extensions to the basal matrix from secondary pigment cells Within the standard cellular composition of the ommatidium in hexapods, the secondary

TE D

pigment cells adjoin the retinula cells and contribute to the basal matrix (Odselius and Elofsson, 1981) at the proximal end of the ommatidium. In both species investigated in this study, no extensions of the secondary pigment cells were found and the SPC are

EP

crammed in between the neighboring PPC, directly beneath the cornea. The reason for the missing extensions is most likely the dense, interlocking, packing of retinula cells and their corresponding nuclei (compare c, d) especially in the distal third of the retinula,

AC C

as found in M. mymaripenne.

4.2.1.3. Interlocking arrangement of the retinula cell nuclei in neighboring ommatidia The interlocking arrangement of retinula cells in neighboring ommatidia, storing their nuclei at different levels, was found in all species, including T. evanescens (Fischer et al., 2011). This illustrates the best use of available space, not only within one ommatidium, but also between neighboring ones. It allows for best packing, when a further decrease in space cannot be achieved by a decrease in size of essential cell organelles, such as the retinula cell nuclei, when these are all at the same height. 4.2.1.4. Distal position of retinula cell nuclei in M. mymaripenne 12

ACCEPTED MANUSCRIPT

As described for T. evanescens (Fischer et al., 2011), and seen now in the two new investigated species, the nuclei of the retinula cells make use of the full length of the retinula cells, with the exception of the very narrow most proximal part. These eyes thus do not show the typical arrangement of nuclei levels known from larger apposition eyes (e.g. Perrelet, 1970), where all nuclei (with the exception of the nucleus of the 9th retinula cell) are located in the distal third of the retinula cells. In M. mymaripenne it is

RI PT

noticeable that part of the nuclei of the regular retinula cells are also shifted in between the primary pigment cells of neighboring ommatidia as a result of the short ommatidium length and the large rhabdom diameter.

SC

4.2.1.5. Elongated shape of the retinula cell pigments

The pigment granules of the retinula cells are not spherical as in larger Hymenoptera (Perrelet, 1970; Skrzipek and Skrzipek, 1974), but ellipsoidal, with the longest extension

M AN U

in a distal-proximal orientation. This allows for a dense pigment sheath, while taking up less space perpendicular to the long axes of ommatidia. The average diameter of the pigment granules is smaller than in larger hymenopterans (Perrelet, 1970; Skrzipek and Skrzipek, 1974), but volumetric measurements are needed in order to validate whether the total volume of the pigment granules is reduced or a volume similar to larger species

TE D

is maintained by this change in their shape. So far a minimal functional and/or structural limit has been postulated (Meyer-Rochow and Reid, 1996), but this has yet to be examined by 3D reconstruction or other methods. Even though the smallest pigment diameters are found in in M. mymaripenne, pigments

EP

are much more densely packed in retinula cells (on average 13 pigment granules per retinula cell in cross-sections of the distal rhabdom of M. mymaripenne (Fig. 2E, 3),

AC C

compared to 5 in A. flavipes (Fig. 4D, E, F)). This reveals the spatial constraints within the retinula cells of M. mymaripenne.

4.2.1.6. Increase of the rhabdom diameter in M. mymaripenne compared with A. flavipes and T. evanescens Insofar as no differences were found in the diameters of the microvilli (60 ± 10 nm) compared with larger hymenopterans (Varela and Porter, 1969; Perrelet, 1970), the shorter length of the rhabdom in the investigated species imposes a loss of lightabsorptive material, the smaller size resulting in fewer microvilli and thus smaller rhabdomeres. However, with decreasing ommatidial length, an increase in rhabdom 13

ACCEPTED MANUSCRIPT

diameter is noticeable in the species investigated. The rhabdom diameter in the distal part in T. evanescens (1.7 µm) and A. flavipes (1.4 µm)

is smaller than in Apis

mellifera (worker honeybee 2 µm, Greiner et al., 2004) or Cataglyphis bicolor (worker ant, approx. 2 µm, Eheim and Wehner, 1971), while the smallest M. mymaripenne maintains a rhabdom diameter of 2.4 µm.

RI PT

Large rhabdom diameters are usually found in nocturnal species, for example up to 8 µm as in Megalopta genalis (Greiner et al., 2004) as an adaptation to low light levels. However M. mymaripenne, as also A. flavipes and T. evanescens, have so far been reported to be diurnally active species (personal observations AP; Quednau, 1958). The ocelli of M. mymaripenne are also of smaller size compared with those of

SC

T. evanescens and A. flavipes. Apart from that, different to larger nocturnal hymenopteran species, which show an increase in ocelli diameter in comparison to their

the investigated species.

M AN U

diurnal relatives (Jander and Jander, 2002; Greiner et al., 2004), this is not the case in

A large diameter of the rhabdom increases the total photon catch, by increasing the total volume of the rhabdomeres as well as the acceptance angle of the rhabdom in

TE D

comparison to species with similar focal lengths, but more slender rhabdoms. Furthermore, rhabdoms with diameters above 2 µm function as lightguides where light is trapped within the rhabdom by internal reflection, whereas rhabdoms with diameters of less than 2 µm function as waveguides (Snyder, 1979, Stavenga, 2003). Due to the

EP

waveguide nature of the rhabdom, part of the propagating light is travelling outside the rhabdom and might be absorbed by the surrounding pigment granules. The larger the

AC C

distance between pigment granules and rhabdom, the more waveguide modes can be propagated Stavenga (Stavenga, 2004). This mechanism is commonly used in the transition between dark and light adaptation in order to adapt apposition eyes to different light sensitivities (e.g. Meyer-Rochow, 1999). It is interesting in this context that the cisternae of the endoplasmatic reticulum found in the eyes A. flavipes (waveguide) are wider compared with those in M. mymaripenne (lightguide).

Three of the described adaptations are also found in similar ways in small lepidopterans (Fischer et al., 2012a,b, 2013), although eye sizes are not that small and are of the optical superposition type. These adaptations can therefore be generalized as a trend in compound eye miniaturization: the decrease in radius of curvature of the facets, 14

ACCEPTED MANUSCRIPT

relatively large rhabdom diameters (the distal rhabdom in case of the lepidopterans) and the size and position of the retinula cell nuclei all have an impact on the morphology of miniaturized ommatidia. Given that we did not find cell reductions in any of the investigated hymenopteran and lepidopteran species (this study, Fischer et al., 2011; Fischer et al., 2013) especially the

RI PT

spatial constraints are of special interest in terms of the limits of miniaturization. Dislocations of essential organelles (e.g. retinula cell nuclei), to allow for compact packing and restricting secondary pigment cells to the most distal position (tiny Hymenoptera only) appear to be adaptations to spatial constraint without imposing functional restrictions. The pigment granules of the SPC´s still prevent stray light in the

SC

distal part of the ommatidium. The elongation of retinula cell pigments is an additional way to deal with the limited space within the cells without losing functionality. However,

M AN U

the dense coverage of the cytoplasm by cell organelles and pigment granules (Fig. 2E, 3B, C) within the retinula cells of M. mymaripenne (on average 13 pigment granules per retinula cell in cross sections versus 5 in A. flavipes), illustrates that there will be limits during miniaturization at which only a reduction in the number of pigment granules and/or the volume of the nuclei can result in yet smaller ommatidia.

TE D

4.3. Functional consequences and limits of compound eye miniaturization For a functional eye, three properties would need to be retained as body size becomes smaller: I) the dioptric apparatus must be able to focus light onto the retinula; II) the

EP

rhabdom itself must be large enough to absorb sufficient photons at typical illumination levels to generate reliable signals; and III) any reduction in cell size must provide sufficient storage for all organelles needed to maintain cellular functionality. All three

AC C

factors are dependent on each other in an eye of given size and one factor can not be optimized without having an impact on the others (Warrant and McIntyre, 1993; Land and Nilsson, 2002). Increasing the dimensions of the rhabdom will, for example, decrease the available space for other cell organelles or decrease the length of the dioptric apparatus. Thus the minimal length of the cone, at which the optics are still able to focus the light on the tip of the rhabdom, will limit the miniaturization of the dioptric apparatus. The optical calculations show that the focal lengths come close to the distal tip of the rhabdom in all species investigated, using approximations for the refractive indices 15

ACCEPTED MANUSCRIPT

taken from real measurements on A. mellifera. We can therefore estimate that a minimal functional limit is not undercut, as long as the real values for the refractive indices do not vary critically from the ones used from A. mellifera. The small outer radius of curvature of the facet allows for a short focal length. A minimal limit for the outer radius of curvature is set roughly by the value representing half of the facet

RI PT

diameter. The short focal length in combination with given rhabdom diameters results in the relatively large acceptance angles of the rhabdoms (4.8 - 22.2°), and these can compensate for shorter rhabdom lengths because more light is channelled to the rhabdom compared with ommatidia having longer focal lengths (and thus smaller

SC

acceptance angles). Still, the sensitivity of A. flavipes (0.04 µm2/sr) and T. evanescens (0.06 µm2/sr), is about half that of the bee Apis mellifera (0.1 µm2/sr; Greiner et al.,

M AN U

2004), while the larger rhabdom diameter (and thus larger acceptance angle of the rhabdom) provides M. mymaripenne with a sensitivity 2.3 times larger than in A. mellifera. Higher values are more typical for nocturnal species or those active in dim light than for diurnal species (e.g. Land and Nilsson, 2002). However, it has to be considered that the formula will overestimate sensitivity under circumstances that apply to the investigated compound eyes (F-numbers < 2, rhabdom diameters < 2 µm), as it

TE D

does not compensate for light loss by rays that will not be trapped in the photoreceptor because of the steep entrance angle (Stavenga, 2003; Warrant and McIntyre, 1991). This can also explain the critical need for a dense cover of pigment granules in the

EP

distal rhabdom of M. mymaripenne in order to absorb stray light. Therefore the real sensitivity of the eye of M. mymaripenne will be closer to the values of diurnal species like Apis, than to dim-light active or nocturnal species. Nevertheless, the sensitivity will

AC C

still be higher in M. mymaripenne compared with T. evanescens and A. flavipes. This sensitivity is yielded by the increased acceptance angle of the rhabdom, which at 22.2o is clearly larger than in the other species (2.7 times larger than for T. evanescens; 4.6 times than for A. flavipes). A comparison with values for the acceptance angle found in other hymenopterans, such as in sahara desert ants (∆ρrh = 4) worker honeybees (∆ρrh =1.7), giant bull ants (∆ρrh =1.7), diurnal halictid bees (∆ρrh =1.6) and yellow sand wasps (∆ρrh = 0.41) (Greiner et al., 2004; Land 1997 (and references cited within)), illustrates the drawback of increasing sensitivity: a severe loss in resolution. But as resolution is not of use at all as long as not a sufficient amount of photons can be obtained to generate a clear signal (Warrant and McIntyre, 1993), sensitivity will be favoured over 16

ACCEPTED MANUSCRIPT

all other parameters during compound eye miniaturization. The need, relatively speaking, to increase the facet diameter in order to adapt the acceptance angle of the rhabdom to the interommatidial angle, can therefore explain the non allometric reduction in the number of facets, which has been shown to decrease in roughly linear proportion to body size in larger species (Yagi and Koyama, 1963; Menzel and Wehner, 1970; Jander and Jander, 2002; Rutowski et al., 2009; Döring and Spaethe, 2009; Fischer et

RI PT

al., 2013). Considering that a further decrease in the focal length (in order to increase sensitivity) is only possible by a further decrease in the outer radius of curvature of the lens (and thus at a certain point smaller facets) stands in contrast to the aforementioned need to have relatively large facet diameters.

SC

In addition to the optical impacts, broader facet diameters also increase the space in the distal part of the retinula cells, relative to equally large ommatidia with similar sized

M AN U

rhabdoms, but smaller facets. This gives the possibility to store pigment granules (which are needed to absorb stray light not entrapped by the rhabdom) and retinula cell nuclei in this region, in addition to an increased rhabdom diameter, as found in M. mymaripenne in which the nuclei even make use of more distal planes. It can be therefore seen as a positive secondary effect during compound eye miniaturization.

TE D

Considering the optical limitations and compromises found in these compound eyes, the question in the following is what these eyes, with low resolution and few facets, as few as 29 in M. mymaripenne, are used for.

EP

All three species are parasitoid wasps, laying their eggs in eggs of other insects (Anderson and Paschke, 1968; Bernardo and Viggiani, 2002; Quednau, 1958).

AC C

Important activities for the females are therefore primarily host detection, mate detection (if applicable, as some hymenopterans are parthenogenetic and males are so far not known for M. mymaripenne) and finding the host eggs in which to oviposit. In terms of the sensitivity of the eyes, the use of vision should be restricted to bright light levels during the day. In a behavioral study T. evanescens was indeed shown to exhibit little to no activity at low illumination levels (Quednau, 1958).

The combination of few ommatidia, large interommatidial angles will not allow details in the landscape to be resolved (compare e.g. Schwarz et al., 2011). However, the compound eyes should be sufficient to allow for flight control (compare e.g. Pix et al. 2000), basic navigation (Schwarz et al. 2011, McLeman et al. 2002) and possibly to 17

ACCEPTED MANUSCRIPT

some extend obstacle avoidance. Resulting from this is, that in general a greater importance on chemical cues and olfaction are to be expected for the tasks such as finding and inspecting of host eggs and mate detection. More complex visual tasks will only be possible with larger eyes having higher resolution, due to larger numbers of facets and larger eye radii and thus smaller interommatidial angles. However, it is important to mention that, before estimating the

RI PT

visual capacities of a species just on the basis of small eyes and few facets, in each case measurements of the internal optical relevant parameters (interommatidial angles etc.) should be performed. The ant Leptothorax albipennis (Curtis 1854), a species with only 60 ommatidia, was shown to be able to use landmarks as reference points

SC

(McLeman et al., 2002). So far nothing is known about the dimensions of the facets or the underlying retina, however this demonstration shows that even complex tasks are possible in compound eyes with relatively few facets. The latter obviously also require

M AN U

an underlying neural processing apparatus of sufficient complexity. 5. Perspective

So far only female specimens of the different species have been compared and it is of interest to mention that the ommatidia of M. mymaripenne are only slighty smaller than

TE D

those of the males of T. evanescens (ommatidium length 24.29 µm; Fischer et al., 2011) although the sexes differ slightly in body size (0.2 mm for female M. mymaripenne versus ≈ 0.3 mm for male T. evanescens). This raises the question about the minimal possible size of an ommatidium in general and about minimal cell sizes and

EP

minimal sizes of nuclei in particular. Gregory (2001) has discussed genome size as a parameter regulating cell and nucleus size, and links between genome size, cell size

AC C

and cell division rates have been identified in different taxa (Gregory, 2002; Gregory and Johnston, 2008). As in other hymenopterans, female Trichogrammatid wasps produce females from fertilized еggs, whereas males develop from unfertilized eggs and are haploid (Heimpel and de Boer, 2008). Thus, if genome size plays a role in miniaturization and the maximum compression of DNA is of importance, differences should exist in nuclear dimensions between males and females of the same species (as long as endoreplication has not taken place in the specific cell types). Size differences are indeed found between the sexes in the compound eyes of T. evanescens with respect to the total length of the ommatidium (female 34.97 µm, male 24.29 µm: Fischer et al., 2011). However, more investigations, including volumetric measurements of cells and nuclei/cell ratios are needed to ascertain whether general correlations can be 18

ACCEPTED MANUSCRIPT

upheld and if a haploid chromosome set allows a more extreme level of miniaturization of cells in compound eyes. These investigations will be the challenge for the future and are currently under preparation by one of the authors (SF). Acknowledgements

RI PT

S.F. gratefully acknowledges the financial support received from the German Science Foundation (DFG) towards this project (FI 1930/1-1). A.M. and A.P. were supported by the Russian Foundation for Basic Research and by the President of Russia Foundation. Comments of two anonymous reviewers helped to improve a previous version of our manuscript.

SC

References

Anderson, R.C., Paschke, J.D, 1968. The biology and ecology of Anaphes flavipes

M AN U

(Hymenoptera: Mymaridae), an exotic egg parasite of the cereal leaf beetle. Annals of the entomological society of America 61, 1-5.

Barlow, H.B., 1952. The size of ommatidia in apposition eyes. Journal of Experimental Biology 29, 667–674.

Bernardo, U., Viggiani, G., 2002. Biological data on Megaphragma amalphitanum and

Megaphragma

Trichogrammatidae),

mymaripenne

TE D

Viggiani

egg-parasitoids

of

Heliothrips

Timberlake

(Hymenoptera:

haemorrhoidalis

(Bouché)

(Thrysanoptera: Thripidae) in southern Italy. Bollettino del Laboratorio di Entomologia

EP

Agraria Filippo Silvestri 58, 77-85.

Döring, T.F., Spaethe, J., 2009. Measurements of eye size and acuity in aphids

AC C

(Hemiptera: Aphididae). Entomologia Generalis 32(2), 77-84. Eheim, W.P., Wehner, R., 1971. Die Sehfelder der zentralen Ommatidien in den Appositionsaugen von Apis mellifica and Cataplyphis bicolor (Apidae, Fromicidae; Hymenoptera) Kybernetik 10(3), 168-179. Fischer, S., Müller, C.H.G., Meyer-Rochow, V.B., 2011. How small can small be: The compound eye of the parasitoid wasp Trichogramma evanescens (Westwood, 1833) (Hymenoptera, Hexapoda), an insect of 0.3- to 0.4-mm total body size. Visual Neuroscience 28, 295-308. Fischer, S., Müller, C.H.G., Meyer-Rochow, V.B., 2012a. Neither apposition nor superposition: the compound eyes of the chestnut leafminer Cameraria ohridella 19

ACCEPTED MANUSCRIPT

Zoomorphology 131, 37-55.

Fischer, S., Meyer-Rochow, V.B., Müller, C.H.G., 2012b. Challenging limits: Ultrastructure and size-related functional constraints of the compound eye of Stigmella microtheriella (Lepidoptera: Nepticulidae). Journal of Morphology 273, 1064-1078. Fischer, S., Meyer-Rochow, V.B., Müller, C.H.G., 2013. Compound eye miniaturization Lepidoptera:

a

comparative

morphological

10.1111/azo.12041

analysis.

Acta

Zoologica

RI PT

in

doi:

Gregory, T.R., 2001. The bigger the C-value, the larger the cell: genome size and red

SC

blood cell size in vertebrates. Blood cells, Molecules, and Diseases 27, 830-843. Gregory, T.R., 2002. A bird´s eye view of the C-value enigma: genome size, cell size,

M AN U

and metabolic rate in the class aves. Evolution 56, 121-130.

Gregory, T.R., Johnston, J.S., 2008. Genome size diversity in the family Drosophilidae. Heredity 101, 228-238.

Greiner, B., Ribi, W.A., Warrant, E.J., 2004. Retinal and optical adaptations for nocturnal vision in the halictid bee Megalopta genalis. Cell and Tissue Research 316,

TE D

377–390.

Heimpel, G.E., de Boer, J.G., 2008. Sex determination in the hymenoptera. Annual Review of Entomology 53, 209-230.

EP

Honkanen, A., Meyer-Rochow, V.B., 2009. The eye of the parthenogenetic and minute moth Ectoedemia argyropeza (Lepidoptera: Nepticulidae). European Journal of

AC C

Entomology 106(4), 619-629.

Jander, U., Jander, R., 2002. Allometry and resolution of bee eyes (Apoidea). Arthropod Structure and Development 30, 179-193. Jenkins, F.A., White, H.E., 1976. Fundamentals of Optics (4th ed.). Auckland, New Zealand: McGraw-Hill. Kunze, P., 1979. Apposition and superposition eyes. In: Autrum H (ed) Handbook of Sensory Physiology, Springer Berlin, Germany VII(6A), 441–502 Land, M.F., 1997.Visual acuity in insects. Annual Review of Entomology 42, 147–77. Land, M.F., Nilsson, D.E., 2002. Animal Eyes. Oxford Animal Biology Series. New York: 20

ACCEPTED MANUSCRIPT

Oxford University Press Inc.

Makarova, A.A., Polilov, A.A., 2013b. The peculiarities of the structure and ultrastructure of the brain of the smallest insects as a result to miniaturization. 2. The smallest hymenoptera (Hymenoptera, Mymaridae, Trichogrammatidae) Entomological Review 92(6), 695-706.

RI PT

McLeman, M.A., Pratt, S.C., Franks, N.R., 2002. Navigation using visual landmarks by the ant Leptothorax albipennis. Insectes Sociaux 49, 203-208.

Menzel, J.G., 1972. The fine structure of the compound eye of Formica polyctena Functional morphology of a hymenopteran eye. In Wehner R (Ed): Information

SC

processing in the visual systems of arthropods. Springer Berlin, Germany, pp. 37-47 Menzel, J.G., Wehner, R., 1970. Augenstrukturen bei verschieden großen Arbeiterinnen

Physiologie 68, 446-449.

M AN U

von Cataglyphis bicolor Fabr. (Formicidae, Hymenoptera). Zeitschrift für vergleichende

Menzel, J.G., Wunderer, H., Stavenga, D.G., 1991. Functional morphology of the divided compound eye of the honeybee drone (Apis mellifera). Tissue and Cell 23(4),

TE D

525-535.

Meyer-Rochow, V.B., 1999. Compound eye: Circadian rhythmicity, illumination, and obscurity. In “Atlas of arthropod sensory receptors–Dynamic morphology in relation to

EP

function”. Ed by Eguchi, E. and Y. Tominaga. Springer. Tokyo. pp. 97–125. Meyer-Rochow, V.B., Gál, J., 2004. Dimensional limits for arthropod eyes with

AC C

superposition optics. Vision Research 44, 2213–2233. Meyer-Rochow, V.B., Reid, W.A., 1996. Does age matter in studying crustacean eye? Journal of Comparative Physiology B 166, 319-324. Müller, R.L., Gregory, T.R., Gregory, S.M., Hsieh, A., Boore, J.L., 2008. Genome size, cell size, and evolution of enucleated erythrocytes in attenuate salamanders. Zoology 111, 218-230. Odselius, R., Elofsson, R., 1981. The basement membrane of the insect and crustacean compound eye: definition, fine structure, and comparative morphology. Cell and Tissue Research 216, 205-214.

21

ACCEPTED MANUSCRIPT

Perrelet, A., 1970. The fine structure of the retina of the retina of the Honey Bee drone. An Electron Microscopical Study. Zeitschrift fur Zellforschung und mikroskopische Anatomie 108, 530-562. Pix, W., Zanker, J.M., Zeil, J., 2000. The optomotor response and spatial resolution of the visual system in male Xenos vesparum (Strepsiptera). The Journal of Experimental

RI PT

Biology 203, 3397–3409. Polilov, A.A., 2007. Miniaturization-related structural features of Mymaridae. In: Studies on hymenopterous insects. KMK, Moscow pp. 50-64. (in Russian).

Polilov, A.A., 2012. The smallest insects evolve anucleate neurons. Arthropod Structure

SC

and Development 41, 29-34.

Polilov, A.A., 2015. Small is Beautiful: Features of the Smallest Insects and Limits to

M AN U

Miniaturization. Annual Review of Entomology 60. (in press).

Quednau, W. (1958). Über einige Orientierungsweisen des Eiparasiten Trichogramma (Hym. Chalcididae) auf Grund von Licht- und Schwerereizen. Anzeiger für Schädlingskunde 31, 83–85.

TE D

Rutowski, R.L., Gislen, L., Warrant, E.J., 2009. Visual acuity and sensitivity increase allometrically with body size in butterflies. Arthropod Structure and Development 38, 9111.

Schwarz, S., Narendra, A., Zeil, J., 2011. The properties of the visual system in the

AC C

134.

EP

Australian desert ant Melophorus bagoti. Arthropod Structure & Development 40, 128-

Skrzipek, K.H., Skrzipek, H., 1974. The ninth retinula cell in the ommatidium of the worker honey bee (Apis mellifica L.). Zeitschrift fur Zellforschung und mikroskopische Anatomie 147, 589–593. Snyder A. W., 1977. Acuity of compound eyes: physical limitations and design. Journal of comparative Physiology 116, 61-182. Snyder A.W., 1979. Physics of vision in compound eyes. In Handbook of Sensory Physiology, Vol. VII/6A, ed. Autrum, H., pp. 225–313. Berlin, Germany: Springer.

22

ACCEPTED MANUSCRIPT

Stavenga D.G., 2003. Angular and spectral sensitivity of fly photoreceptors.II. Dependence on facet lens F-number and rhabdomere type in Drosophila. Journal of Comparative Physiology A 189, 189–202. Stavenga D.G., 2004. Angular and spectral sensitivity of fly photoreceptors. III.Dependence on the pupil mechanism in the blowfly Calliphora. Journal of

RI PT

Comparative Physiology A 190, 115-129. Yagi, N., Koyama, N., 1963. The compound eye of lepidoptera: approach from organic evolution. Shinkyo Press, Tokyo, p 319

Varela, F.G., Porter, K.R., 1969. Fine structure of the visual system of the Honeybee

SC

(Apis mellifera) I. The retina. Journal of Ultrastructure Research 29, 236-259.

Varela, F.G., Wiitanen, W., 1970. The optics of the compound eye of the honeybee

M AN U

(Apis mellifera). The journal of general Physiology 55, 336-358.

Warrant, E.J., McIntyre, P.D., 1990. Limitations to resolution in superposition eyes. Journal of comparative Physiology A 167, 785-803.

Warrant, E.J., McIntyre, P.D., 1991. Strategies for retinal design in arthropod eyes of

TE D

low F-number. Journal of omparative Physiology A 168, 499-512. Warrant, E.J., McIntyre, P.D., 1993. Arthropod eye design and the physical limits to spatial resolving power. Progress in Neurobiology 40, 413–461.

EP

Warrant, E.J., Nilsson, D.E., 1998. Absorption of white light in photoreceptors. Vision

AC C

Research 38, 195–207.

Figure captions

Fig.1. Scanning electron micrographs of the eyes of two tiny parasitoid wasp species. Overall appearance of the compound eyes of M. mymaripenne (A) and A. flavipes (C). Detail of the compound eye, showing the facets and interfacetal hairs of M. mymaripenne (B) and A. flavipes (D). OC – ocelli. Fig. 2. Transmission electron micrographs showing the ultrastructure of the eyes of M. mymaripenne. A - General view of a cross section through the compound eye; B – longitudinal section showing the dioptric apparatus; C – cross section at the level of the 23

ACCEPTED MANUSCRIPT

cone cells, D – semi-schematic drawing of an ommatidium; E – cross section of the distal part of the rhabdom, the duplets of two rhabdomers with microvilli orientated in the same direction each, are separated by the four cone cell projections (CP); F – cross section of the proximal part of the rhabdom; G – longitudinal section through the ommatidium, H – longitudinal section of the PPC. BM – basal matrix; CC – cone cell; CER – cerebrum; COR – cornea; CP – cone cell projections; NC – nucleus; PG –

RI PT

pigment granule; PPC – primary pigment cell; NCSPC – nucleus of the secondary pigment cell; NCPPC – nucleus of the primary pigment cell; RBD – rhabdom; RETC – retinula cell; SPC – secondary pigment cell.

Fig. 3. Transmission electron micrographs of cross section of the ommatidia. Showing

SC

the position of the 9th retinula cell of A – A. flavipes, B – M. mymaripenne; C – revealing the interlocking position of nuclei of the neighboring retinula cells of M. mymaripenne.

M AN U

CC – cone cell; CP – cone cell projection; MT – mitochondria; NC – nucleus; R1-R9 – retinula cells; PG – pigment granule; PPC – primary pigment cell; RBD – rhabdom; RETC – retinula cell.

Fig. 4. Transmission electron micrographs showing the ultrastructure of the eyes of A. flavipes.

A – General view of a cross section through the compound eye; B –

TE D

longitudinal section showing the dioptric apparatus; cross section through C – cone cells; D – the complete compound eye; E – distal part of rhabdom; F – proximal part of the rhabdom; longitudinal section through the G – compound eye, H – proximal part of the rhabdom and basal matrix. ANT – antennae; BM – basal matrix; CC – cone cell;

EP

CER – cerebrum; COR – cornea; CP – cone projections; NC – nucleus; PG – pigment granule; PPC – primary pigment cells; NCSPC – nucleus of the secondary pigment cell;

AC C

NCPPC – nucleus of the primary pigment cell; RBD – rhabdom; RETC – retinula cells; SPC – secondary pigment cells.

Formulas:

Thick lens formula:

24

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

sensitivity formula:

TE D

distance to the vertex of the back surface of the lens

25

ACCEPTED MANUSCRIPT

Table 1. Measurements of external and internal features of the compound eyes of female specimens of Anaphes flavipes and Megaphragma mymaripenne presented in comparison to data available for females of Trichogramma evanescens (Fischer et al., 2011). Data are expressed as means ± s.d.; n = 10. Parameter

Unit

Anaphes

Trichogramma

Megaphragma

flavipes

evanescens

mymaripenne

RI PT

(Fischer et al.2011)

mm

0.45

0.3-0.4

0.2

eye height

µm

97.2 ± 3.1

71.11 ± 3.52

52.6 ± 2.6

eye width

µm

75.9 ± 3.4

59.09 ± 2.76

30.7 ± 1.4

number of facets

121 ± 14

SC

body size

128 ± 6.51

29 ± 1

42.82 ± 0.92

18.2 ± 2.0

4.59 ± 0.27

3.4 ± 0.3

µm

41.8 ± 4.0

radius curvature cornea (outside)

µm

5.9 ± 0.7

radius curvature cornea (inside)

µm

5.0 ± 3.6

6.39 ± 0.82

2.8 ± 0.2

facet diameter

µm

8.1 ± 0.6

6.39 ± 0.33

8.1 ± 0.3

length ommatidium

µm

33.4 ± 8.6

34.97 ± 3.88

22.7 ± 0.9

interommatidial angle

deg.

15.0 ± 0.6

9.98 ± 1.57

21.5 ± 0.8

length dioptric apparatus

µm

13.4 ± 2.7

10.26 ± 0.98

7.4 ± 0.6

maximum thickness cornea

µm

3.5 ± 0.7

2.25 ± 0.29

2.6 ± 0.4

thickness interfacet cornea

µm

1.0 ± 0.3

not measured

0.3 ± 0.04

TE D

cone length

M AN U

radius curvature eye

µm

10.0 ± 0.7

7.88 ± 0.71

5.3 ± 0.4

µm

7.1 ± 0.8

5.51 ± 0.52

5.2 ± 0,4

µm

19.8 ± 0.9

not measured

13.3 ± 0.5

µm

1.4 ± 0.2

1.67 ± 0.12

2.4 ± 0.4

diameter of rhabdom microvilli

nm

72.9 ± 2.6

62.22 ± 4.02

57.4 ± 3.3

pigment granule diameter (PPC) and length*

µm

0.44 ± 0.04 0.79 ± 0.08*

0.44 ± 0.04

0.45 ± 0.1

pigment granule diameter (SPC) and length*

µm

0.42 ± 0.03 0.75 ± 0.08*

0.33 ± 0.03

0.28 ± 0.05

pigment granule diameter (retinula cells) and length*

µm

0.23 ± 0.03 0.69 ± 0.04*

0.28 ± 0.04

0.28 ± 0.04 0.73 ± 0.08*

5±2

not measured

13 ± 5

maximum cone diameter length rhabdom

AC C

EP

rhabdom diameter (distal end)

pigment granule density (number retinula cell in cross sections)

per

diameter of nuclei of cone cells

µm

2.0 ± 0.1

not measured

1.9 ± 0.3

diameter of nuclei of PPC

µm

1.5 ± 0.5

not measured

1.2 ± 0.3

diameter of nuclei of SPC

µm

0.9 ± 0.2

not measured

1.5 ± 0.3

diameter of nuclei of retinula cells

µm

1.8 ± 0.4

not measured

1.2 ± 0.3

14.2 ± 0.64

not measured

2.2 ± 0.32

number of interfacetal hairs thickness of basal matrix

µm

0.1 ± 0.02

not measured

0.2 ± 0.04

ocelli diameter

µm

8 and 12.5 ± 3.8

11 and 15

7.1 ± 0.6

ACCEPTED MANUSCRIPT

*Due to oval shape of the pigment granules it is necessary to measure both axes:

AC C

EP

TE D

M AN U

SC

RI PT

diameter and length of granules.

ACCEPTED MANUSCRIPT

Table 2. Optical parameters calculated from eye measurements of of M. mymaripenne and A. flavipes.

f´

∆ρrh

∆ρl

S

(µm)

(µm)

(deg.)

(deg.)

(µm2/sr)

M. mymaripenne

6.2

8.4

0.8

22.2

3.5

0.23

T. evanescens

8.9

12.0

1.4

8.1

4.5

0.06*

A. flavipes

10.7

14.4

1.3

4.8

3.5

0.04

SC

F-number

RI PT

f

AC C

EP

TE D

resulted in a lower value (0.04).

M AN U

*transposed digits in the calculation for the original publication (Fischer et al. 2011)

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

The eye ultrastructure of one of the smallest known hymenoptera was investigated for the first time.

AC C

EP

TE D

M AN U

SC

RI PT

Size-related adaptations in compound eyes were detected and discussed.