With multiple independent researchers contributing knowledge to a taxonomic group, communities tend to adopt the same views and formats to better communicate with each other. Scientific illustrators are taught to be aware of conventions operating within the scientific discipline to which they are contributing. “Maintaining consistent conventions permits the work of several illustrators to be easily compared and ensures that an illustration will be ‘read’ properly” (Hodges 2003). In digital imaging, parameters such as lightning, optical, and specimen orientation are kept consistent. Distributed collaborative projects such as AntWeb (2017) have explicit standards and instructions for creating digital images of standard views (Nelson et al. 2012). When executed according to the protocol, images and data contributed to the site will be comparable regardless of the supplier (see antweb.org/documentation.do). Standard formats are used to facilitate the transfer and sharing of data (Haston et al. 2012, Wieczorek et al. 2012). Standards in scientific imaging minimize creative variation to ensure that the subject is represented in a consistent way and can be integrated into the corpus of scientific literature. Because of the need to comply with standards, we argue that such images lack “sufficient individuality”, the central criterion used to determine if an illustration qualifies as a “work” in the sense of copyright law.

The combination of structured text and standard view images in taxonomic treatments is a mechanism for documenting facts (Winston 1999). The approach is not expressive in the sense that creative writing and visual arts are. This contrasts with other representations of natural history, such as wildlife illustrations created as pieces of commercial art (Hodges 2003), and also with examples of expressive creativity that occasionally appear in scientific literature (Fig. 1). Such works are excluded from the rights arguments made here.

Taxonomy began as a scientific discipline in the middle of the 18th century. Botany and zoology designate different works of Carl Linnaeus as the starting points for their respective taxonomic domains: Species Plantarum (Linnaeus 1753) for botany, and the 10th edition of Systema Naturae (Linnaeus 1758) for zoology. Both publications include a standard naming system, a hierarchical classification system, taxonomic treatments reporting key characteristics and distribution, and citations of earlier publications. The value of drawings was not initially grasped and early works such as those of Linnaeus (Linnaeus 1753, Linnaeus 1758) and the pioneer protistologists Otto Müller (1773) lacked illustrations. The earliest illustrations recognized by zoological taxonomy appear in Aranei Svecici, a 1757 publication on the spiders of Sweden by C.A. Clerck (1757). Although actually published before the official start of zoological nomenclature, Aranei Svecici is explicitly recognized by the International Code of Zoological Nomenclature (International_Commission_on_Zoological_Nomenclature 1999: Art. 3). Aranei Svecici contains illustrations of nearly 70 spider species. Nearly all of these feature a full body illustration (habitus) showing the dorsal view with legs symmetrically arranged. In a few cases, the male intromittent organ (the pedipalp) is illustrated. In many taxonomic groups including spiders, reproductive structures are rich in complex characters that show consistency within and differences between species (Eberhard 1985, Song and Bucheli 2010). This makes reproductive structures valuable for recognizing and classifying many taxa, and they are frequently depicted in taxonomic treatments. Pedipalps are a pair of leg-like appendages that arise from the anterior part of the spider. As such, a pedipalp can be positioned and viewed in a limited number of cardinal orientations. When extended straight ahead and rotated in a transverse plane, four anatomically significant views are exposed in increments of 90°: dorsal, ventral, and two lateral views. In Aranei Svecici, the illustrations of the genitalia are less consistent than the habitus, but all have the pedipalp oriented along a cardinal anatomical axis. In the case of the green huntsman spider Micrommata virescens, both the habitus and male pedipalp are included (Fig. 2). The left pedipalp is illustrated from the left side (the retrolateral view; 180° from the prolateral view). A more recent guide to the spiders of Great Britain and Ireland (Roberts 1985) includes illustrations of both the habitus and pedipalp in the same orientation as Clerk first depicted in Aranei Svecici. Unsurprisingly, the more contemporary examples are more detailed and accurate, and the orientation of the images on a page may differ. Nonetheless, the selection of what to illustrate and how to orient it are unchanged despite the nearly 230 years separating these two publications.

Comparative anatomy is a dominant organizing principle in taxonomic publications, regardless of the domain of life concerned. Fig. 3 shows illustrations of Parnassia palustris flower anatomy from Linnaeus to the late 20th century. Key structural features are consistently visible across this time series. Similarly, a series of 18th and 19th century illustrations of the false chanterelle mushroom Hygrophoropsis aurantiaca show the same developmental stages and highlight the same anatomical details (Fig. 4).

Foraminifera are single-celled amoeboid protists, mostly less than 1 mm in length, that typically construct a test (or shell). Although Foraminifera are relatively simple organisms, the cardinal orientations of their anatomy have long been recognized by taxonomists. Fig. 5 includes excerpts from three taxonomic publications that deal with Sigmolina sigmoidea. It resembles a compressed sphere with a c-shaped pore at one end. A study from 1884 (Brady 1884) and another from 1917 (Cushman 1917) depict this species with the same three standard views: a lateral view, a straight on view centered on the aperture, and an axial cross section. Another work from 1974 (Ponder 1974) depicts several Sigmolina species, but employs the same three standard views to depict and compare them. Despite the variety of forms in the axial cross section view of several species, the standard view makes them comparable.

Scientific illustration can be expensive and time consuming to prepare, and costly to publish. This has historically placed limits on how thoroughly a treatment can be illustrated. For example, Biologia Centrali Americana (1879-1915) was a massive effort to document a regional fauna. It comprised 63 weighty volumes and included 1677 figure plates. But only 37% of the species treated were illustrated, and most of those species that were illustrated appeared in only one or two figures (Ramírez et al. 2007). Nevertheless, illustrations were generally limited to a few standard views. As in the previous examples, cardinal directions guide orientation. Fig. 6 compares a plate from the first Biologia Centrali Americana volume on the insect order Orthoptera (grasshoppers, katydids, and their allies) (Saussure 1893) to Naskrecki’s more recent book on the Katydids of Costa Rica (Naskrecki 2000). Both sources include a habitus in lateral view, habitus in dorsal view (which may be only partial), multiple views of the head region, and genitalia. Like most contemporary taxonomists, Naskrecki (2000) depicts a core of standard views for all the taxa treated to facilitate comparison. The Orthoptera volumes of Biologia Centrali Americana depict many of the same standard views. But because many species are not illustrated for most standard views, there are gaps in knowledge that can make it difficult to apply Biologia Centrali Americana as a taxonomic guide.

Interpretive difficulties arise when images of the same structure are not illustrated in the same way or from the same angle (Ramírez et al. 2007). In an example from spider taxonomy, a 1942 publication by Chamberlin and Ivie (1942) included treatments of nearly all known Linyphantes (Arachnida: Araneae: Linyphiidae) species, but did not include illustrations of the pedipalp in any commonly used orientation. The apical view is useful for distinguishing Linyphantes species from each other, but without also including widely used standard views, it is difficult to compare Linyphantes to other genera, such as Bathyphantes. In 1929, Petrunkevitch (1929) published the only reference to include illustrations (albeit rudimentary) of both the retrolateral and apical views together (Fig. 7).

As taxonomic knowledge within any particular group grows, community consensus about the relative value of various standard view images evolves. The importance of standard views to facilitate comparison has remained unchanged even as technologies and techniques have evolved, facilitating the inclusion of more numerous, high-quality images.

Taxonomists and scientific illustrators use a variety of media to capture and convey the morphology and anatomy of organisms. Traditional techniques apply ink, graphite, paint, or other such substances alone or in combination to paper, board, or other such surfaces (Hodges 2003). For most of the history of taxonomy, line drawings (black ink on paper) have been the most widely used medium for depicting anatomy, complemented by colored plates and, in the early 20th century, photography.

Starting around the mid-1980s, new technologies introduced alternative mechanisms for capturing and rendering information about morphology. Computer-aided illustration techniques were developed. Mixed media approaches made it possible to combine multiple techniques into single composite images, such as a body rendered by hand in pencil combined with photographs of wings (Fig. 8). The increasing availability of scanning electron microscopy (Fig. 8) opened new frontiers of discovery (Claugher 1990, Coddington 1989, Stant 1973). Advances in digital cameras mounted on microscopes, and the advent of extended focus composite imaging (Buffington et al. 2005, Riedel 2005) reduced the time and expense of graphically representing morphology (compared to illustration in particular), and photographs began to eclipse illustrations as the primary means of documenting morphological structures (Fig. 9; e.g., Riedel et al. 2013).

Other radiation imaging techniques, such as X-rays, are used to detail skeletal elements in animals, and tomography (micro-CT, synchrotron) is increasingly used to compare detailed anatomy with the aid of three dimensional computer models (Fig. 10); Faulwetter et al. 2013). With three dimensional interactivity, structures can be compared from any angle. Sonograms are used to represent and compare the sounds made by organisms such as birds, crickets, bats, and whales (Fig. 11).

Taxonomic publications often feature photographs from the field, typically depicting living organisms and their habitats (Fig. 12). Such photographs may not be structured and standardized with the precision of a standard view anatomical illustration, but the purpose of such photographs is to document facts, such as color and posture of the organism in life, habitats where it has been found, behavior and interactions with others, and more. Aesthetic and artistic considerations are secondary.

Various automated methods are increasing used to capture information. Camera traps are automated image capture systems left in the field for an extended time to document wildlife activity in a particular location (O'Connell et al. 2011; Fig. 13). Camera trap images rarely feature in true taxonomic publications, but contribute knowledge about where a particular species occurs, and thus provide observation data for scientific publications and conservation management (Griffiths and van Schaik 1993, Karanth 1995, Silveira et al. 2003). Other automated techniques included mass-digitization of museum and herbarium specimens (Blagoderov et al. 2012, Haston et al. 2012, Holovachov et al. 2014, Nelson et al. 2012); Fig. 14), robotic imaging of the sea floor or other inaccessible habitats (Kwasnitschka et al. 2016, Silveira et al. 2003), and flow cytometers with automatic image capture to take pictures of phytoplankton (Dubelaar and Jonker 2000, Sieracki et al. 1995).

Of special significance to taxonomists are images of specimens that are held in institutions such as museums and herbaria. These are estimated to be over 3 billion specimens in about 55,000 museums and 3,500 herbaria around the world. Many are of individual organisms that are significant to the taxonomic or nomenclatural history of the taxon. Of these, the most important are the organisms that are type material as they help establish the identity of taxa. All other specimens help to clarify variation within and among species. As taxonomists need to inspect the materials on which nomenclatural and taxonomic decisions are made, they require access to the preserved material. Historically, taxonomists had to visit repositories or have materials shipped to them. This was costly and specimens were at high risk of being damaged if shipped around the world. Now the use of high resolution images inclusive of 3D images is effective for most purposes, cheap and with low risks of damage. This has led to the investment in specimen digitization programs, such as iDigBio is the US (Nelson et al. 2012). Taxonomic materials are presented using standard techniques, such as pinned insects or herbarium sheets for plants (Bridson and Forman 1998, Häuser et al. 2005, Leenhouts 1968, Metsger and Byers 1999, Schauff 1986).

Taxonomic publications often include maps, typically to show the geographic distribution of occurrence records. Base maps may be static, from a printed or graphical source, or rendered as a layer in a GIS (Geographic Information System) environment. Google allows annotation and non-commercial distribution of its maps including their use in journal articles when proper attribution is provided (Fig. 15; google.com/permissions/geoguidelines.html).

Individual figures are often combined to form plates composed of several species to facilitate comparison (Fig. 16). The images may be arranged to represent relative size (with larger and smaller subjects shown at a common scale), or at different scales with a scale bar included to insure that actual size can be inferred. Such plates are especially common in field guides, where the primary purpose is efficient taxonomic determination.

Quantitative data may be represented as scatter plots (with or without trend-lines), graphs, histograms, pie charts, and other such devices. Charts provided for the purpose of establishing or comparing the distinguishing characteristics of a species lack the requisite creative element that makes copyright applicable (Fig. 17).

To accommodate the growing collection of digitized images relevant to taxonomy, online image archives are growing. These include, Morphbank, BHL on Flikr, and the Biodiversity Literature Repository, which at the time of writing includes over 100,000 images extracted from 15,000 taxonomic publications. Image analysis is being used to automatically identify specimens based on libraries of reference images (Santana et al. 2014). New tools are integrating images and distribution data with mobile phone technology to facilitate field identification taking into account the users current location and proximity to records of various species (e.g., Map of Life, eBird). This exemplifies the fundamental purpose of images in taxonomy: references for identification of known biodiversity.

The criteria for determining whether copyright applies to a class of images are the same regardless of subject matter. We emphasize here those classes of images most applicable to taxonomy, but the principle applies to other domains of science. That is, if the image adopts conventions intended to facilitate comparison with other works, then the image is unlikely to be a creative work in the sense of copyright. This does not mean that images in taxonomy are less important than those from creative fields, only that copyright protection is neither applicable legally nor desirable in the context of comparative science.

National laws specify which non-tangible goods may be regarded as intellectual property. As is the case with property rights with respect to tangible goods, each country may have different rules for intellectual property rights. International instruments such as treaties and conventions aim to harmonize national rules and reduce discrepancies by fixing minimum standards and recommending rules for the application of rights. Various international instruments address specific branches of intellectual property. Well known examples include the Berne Convention for the Protection of Literary and Artistic Works (World Intellectual Property Organization 1979), the Rome Convention for the Protection of Performers, Producers of Phonograms and Broadcasting Organizations (World Intellectual Property Organization 1961), and the Paris Convention for the Protection of Industrial Property (World Intellectual Property Organization 1979b). These international instruments apply only to the extent that they are represented in national laws.

The protection of non-tangible goods is always limited to specific areas and specific objects; there is no "general protection". If national laws do not specify that particular non-tangible goods are objects of intellectual property rights, then no rights apply. Individuals cannot claim intellectual property rights over items that are not covered by the relevant national laws.

Intellectual property rights with respect to non-tangible goods is always limited to a restricted number (the so-called "numerus clausus") of specifically attributed rights (Troller 1959). With regard to scientific images, there are only four relevant areas of intellectual property rights: copyright (or “authors' rights”, as it is referred to in international conventions and in most European countries), the EU-specific database protection, protection against unfair competition, and, in a few European countries, special protection for photographs. We discuss these areas below.

Copyright protects certain human creations in art and literature. The minimum standard of this protection, applicable within the 172 countries that have introduced this form of intellectual property rights into their national laws, is defined in the Berne Convention for the Protection of Literary and Artistic works. The Convention was first established in 1886 and has been amended several times (World Intellectual Property Organization 1979a). It has become the international minimum standard for copyright protection through the implementation into the Agreement on Trade-Related Aspects of Intellectual Property Rights which forms Annex 1C to the Agreement establishing the World Trade Organization in 1994 (World Trade Organization 1994aBeier and Schricker 1996).

Article 2 paragraph 1 of the Berne Convention (World Intellectual Property Organization 1979a) declares that "the expression 'literary and artistic works' shall include every production in the literary, scientific and artistic domain, whatever may be the mode or form of its expression, such as books, pamphlets and other writings; (...) works of drawing, painting, architecture, sculpture, engraving and lithography; photographic works to which are assimilated works expressed by a process analogous to photography (...)." Member countries of the Berne Convention are therefore obliged to protect illustrations such as “drawings” and other artistic works or "photographic works" by their national copyright law (Masouyé 1981).

Copyright confers to the author a set of privileges which result in far-reaching control over access to the work and over most forms of re-utilization. These rights are limited in time (normally to 50 or 70 years after the author’s death) and may be restricted by numerous legally defined exceptions and limitations. Again, there are important differences from country to country with respect to the content and the extension of rights conferred to authors as well as to the definitions of exceptions and limitations. In the United States of America, the “fair-use-principle” (see below "Exceptions and limitation, fair use") substitutes for the concept of exceptions and limitations.

When a concept or intent is captured in the form of a photograph, graph, drawing or illustration, it is said to be ‘fixed’ or ‘expressed’. The Berne Convention (World Intellectual Property Organization 1979a: Art. 2 paragraph 2) allows the member countries “to prescribe that works in general or in any specified categories of works shall not be protected unless they have been fixed in some material form”. This voluntary restriction was confirmed by the 1996 WIPO Copyright Treaty (Reinbothe and von Lewinski 2002). It exists in the United States of America and in many other jurisdictions.

The Berne Convention does not define the notion of "work", but leaves that definition to national legislators. As a consequence, the notion of “artistic work” or "photographic work" can vary from country to country. However, there are aspects of this term that are identical in all copyright systems. One of them is that “works” must be man-made. Objects produced by nature or by organisms never qualify as copyrightable. Another important criterion is that intellectual productions qualify as works only if they are somehow original. This criterion does not refer to the content of the work, but to the form of presentation (Masouyé 1981: section 2.8). Copyright applies to a "work" only if it is expressed in an original (individual, new, creative) way (Lucas and Lucas 1994). In the case of a photographic picture, it can only be considered as a copyrightable photographic work if it somehow differs in style compared to other photographs taken of the same object or other similar photographs with which it may be compared. An interesting example is the recent ruling of a Swiss Court denying copyright protection to a photograph showing a well-known panoramic view of the city of Basel despite the huge technical effort made in producing the photo (Appellationsgericht des Kantons Basel-Stadt, Besondere zivilrechtliche Abteilung 2016).

The same applies to drawings and other forms of scientific illustrations: they are works in the sense of copyright if they adopt an original form of expression. Illustrations that follow predefined rules or conventions do not qualify as copyrightable works. Illustrations of biological information, especially in taxonomy, usually follow conventions that facilitate comparisons with similar illustrations. When this is the case, the images do not qualify as copyrightable works.

According to U.S. Copyright Law, a work may not qualify for copyright protection if it is about an "idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work." (Legal Information Institute 1976a: 17 U.S. Code § 102 (b)). The paragraph describes a concept that is basic for all copyright laws in the world (Masouyé 1981: section 2.8). It has been expressly retaken in Art. 9 par. 2 of the TRIPs-Agreement (World Trade Organization 1994).

The mechanical reproduction of an already existing photograph, drawing, painting or other forms of two-dimensional presentation (such as herbarium sheets) cannot qualify as photographic works for copyright purposes (Heitland 1995). Objects of photographic works must be three-dimensional. If the two-dimensional object of the photograph is a copyrightable work, the photographs qualify as reproductions of the copyrighted work, but are not photographic works in themselves.

Copyright protection, does not only refer to single works, but also to collections of objects (Art. 2 paragraph 5, Berne Convention). Again, the qualification as copyrightable work requires that there is originality and individuality by reason of the selection or the arrangement of the objects. This condition is expressly stated in Art. 10 paragraph 2 of the TRIPs-Agreement (World Trade Organization 1994a) as well as in Art. 5 of the WIPO Copyright Treaty (World intellectual Property Organization 1996). As established by the U.S. Supreme Court decision in Feist Publications, Inc. v. Rural Telephone Service Co. (Legal Information Institute 1991: 499 U. S. 340), collections of objects that are presented in a standardized form, for example in alphabetical order, or in the case of Biology following a predefined systematic order, will not qualify as copyrightable works. The inclusion of single drawings within a plate to combine, summarize or compare attributes of organisms also follows established conventions and such plates are therefore ineligible as copyrightable works.

In most countries, databases are protected by intellectual property rights to the extent that they qualify as works in the sense of copyright. This is the case where there is individuality in the selection of data or in the form of presentation of these data. Databases that do not meet these requirements are not subject to specific protection rules.

An important exception to this rule exists in the E.U. The E.U. introduced, with directive 96/9/EC (The European Parliament and the Council of the European Union 1996), a special protection for databases that is independent of, and in certain cases complementary to, copyright protection. This so-called sui generis protection applies to databases which show “that there has been quantitatively and/or qualitatively a substantial investment in either the obtaining, verification or presentation of the contents” (Art. 7, directive 96/9/EC). This allows the person who invested in the creation of the database to prevent the extraction or re-utilisation of the whole or a substantial part of the contents of that database.

The term "database" is defined in Art. 1 no. 2 of directive 96/9/EC: "For the purposes of this Directive, 'database' shall mean a collection of independent works, data or other materials arranged in a systematic or methodical way and individually accessible by electronic or other means." This is consistent with a data environment being structured into one or more tables, of tables having one or more fields, and fields holding data. The fields are defined by metadata. The content of such databases is made visible or can be copied using web-services or web interfaces.

Database protection does not deal with individual data elements. The intellectual property right refers to the database as a whole, not to an individual datum. Database protection may therefore apply to a collection of scientific images, but not to an individual image. The protection is very specific to prevent the extraction and re-utilization of the database as a whole or of substantial parts of it. It does not serve to prevent the extraction and re-utilization of individual data or of groups of datasets that do not constitute a substantial part of a database.

As the European Court of Justice has pointed out in several judgments, European database protection concerns the creation of databases out of material that already exists, but does not deal with the creation of data as such. “Investment in the obtaining, verification or presentation of the contents” refers therefore to the resources and efforts that were called on to find, collect, verify and/or present existing materials. What constitutes a ‘substantial investment’ was explored in case (InfoCuria - Case-law of the Court of Justice 2004) in which the British Horseracing Board (and others) had objected to the re-use of the content of their database. Their case failed as the Court estimated that the collection and the presentations of the horseracing previsions and results did not require a substantial investment and, in consequence, the extraction and reuse of data was regarded as not being in contravention of database protection. This case is relevant to biology as many databases take pre-existing digital information from other sources and organize the data using widely accepted standard metadata, ontologies, and identifiers. Increasingly, biodiversity-oriented data environments (such as Catalogue of Life, Global Biodiversity Information Facility, Biodiversity Heritage Library, International Plant Name Index, Encyclopedia of Life, or Ocean Biogeographic Information Service) rely to some extent on the content contributed by other databases or by individuals, projects and organizations. Such databases are likely to be ineligible for database protection and the use of some of the content of European biodiversity databases is likely to be legitimate. The value of such databases lies not in their content, but on the extent to which they are maintained to be current and accurate.

Many countries have legal regulations which seek to prevent unfair competition in industrial and commercial matters. The minimum standard for this protection, applicable in 194 countries, is defined by the Paris Convention for the Protection of Industrial Property, established in 1883 and amended most recently in 1979 (World Intellectual Property Organization 1979a). Article 10bis of the convention defines as prohibited unfair competition:

1. all acts of such a nature as to create confusion by any means whatever with the establishment, the goods, or the industrial or commercial activities, of a competitor;

2. false allegations in the course of trade of such a nature as to discredit the establishment, the goods, or the industrial or commercial activities, of a competitor;

3. indications or allegations the use of which in the course of trade is liable to mislead the public as to the nature, the manufacturing process, the characteristics, the suitability for their purpose, or the quantity, of the goods.“

Many countries consider that one form of unfair competition is to reproduce and exploit a competitor’s product or service which is ready for marketing without contributing any novel performance or investment. This legal protection does not aim at a defined intellectual property right, but at lawful commerce. It’s actions prevent behavior that could harm fair competition in an open market.

With respect to scientific images, it might constitute unfair competition to reproduce published images and sell them for individual profit. Unfair competition protection only applies if there is competition between the publisher of the images and the seller. The competition law does not prevent the utilization of published images for other non-competing purposes, such as for any scientific use.

A few European countries such as Germany and Austria have introduced special protection for photographs. The purpose is to protect against unfair competition. Photographers in these countries have a specific intellectual property right in their photographic production, but it applies only within that country. The right lasts for 50 years from the date of publication and protects against every form of re-use.

This special protection must be understood in the light of its historical background (Dommann 2014). A revision of the Berne Convention was to take place in 1908 in Berlin. France asked for the extension of copyright protection to photographs. The German Reich was strictly opposed to this petition as it feared negative effects for its growing photographic industry. In order to prevent the French proposal, the German Reich introduced in 1907 this special protection for photographs granting the photographers fewer prerogatives than a copyright and lasting only 25 years. The Conference in 1908 ended with a compromise agreement that both solutions - copyright on one side, special protection on the other side - were acceptable in light of the Berne Convention. The result was that Germany did not protect photographs through copyright law. At the dawn of World War II, some countries under German influence (Austria, Denmark, Italy) followed their example. In 1948, the Berne Convention was revised again and at this time the copyright protection of photographs became compulsory. Instead of replacing the special protection with copyright protection, the aforementioned countries introduced the copyright protection for photographs into their national law, but also maintained the former protection. This double protection, referring to different kinds of photographs, was upheld also in later law revisions.

The specific photograph protection applies only to non-individual photographs, taken from three-dimensional objects. As it is the case in copyright law, the reprography of a print, a drawing or a pre-existing photography is not a photograph in the sense of these laws (Schricker and Loewenheim 2010: N. 22 zu § 72 UrhG). The protection is rather difficult to apply and has only little importance in practice. However, researchers working in one of these countries should be aware that the re-use of photographs under these legal systems is more problematic than in the rest of the world. Researchers not based in these countries, but wanting to use photographs from these countries, are not subject to this restriction.

Copyrightable works are defined as individual, original human creations, that show originality in the form of presentation compared to other works of the same kind. Most scientific images lack an original form of expression and so cannot qualify as copyrightable works. This is particularly true for machine-generated images, such as robotic systems used to digitize specimens in natural history collections, or pictures obtained from camera-traps positioned to monitor animal activity over time. Such pictures are not man-made and they can consequently not be copyrightable works. For the same reason, they do not qualify for the special photograph protection that applies within a few European countries.

Individually prepared photos and drawings, produced in line with scientifically recognized and standardized conventions, also fall outside the scope of copyright because of their standardized form of expression (Appellationsgericht des Kantons Basel-Stadt, Besondere zivilrechtliche Abteilung 2016). Routine photographs and scans made from two-dimensional objects, as for example photos of herbarium sheets, are not copyrightable as they lack individuality and creativity (Fig. 14a).

Similar arguments apply to the combination of text and standardized images that make up taxonomic treatments. Treatments follow conventions to facilitate the effective documentation of facts, and comparison between descriptions. The expectations are so firm that peer review would not allow treatments that are individual in the sense of literature or art. They are technically “correct” if they are done according to the applicable protocols, and they are “incorrect” if they do not follow those standards. They express facts in a pre-established, standardized form. They do not, therefore, qualify as copyrightable works (Agosti and Egloff 2009).

The same criterion leads to the conclusion that collections of biodiversity data are also not copyrightable by default as the selection and arrangements of those collections as well as their form of expression follow predefined systems of biological classification, metadata, ontology, vocabulary and quantitative units. Tables of quantitative or qualitative information can be considered as collections of data, the selection and presentation of which are scientifically predefined. The more complete and systematic a collection, the less probable it is that it qualifies as a work in the sense of copyright. This conclusion does not devalue scientific work, but it is a logical consequence of copyright legislation that aims to protect individual forms of expression.

The situation is less consistent as far as wildlife illustrations are concerned. Some images are created for artistic purpose or to create a commercial product. Some photographs or drawings generated during field research and which are not produced in line with established standards, may fulfil the criterion of individuality and originality and therefore qualify as works in the sense of copyright. Copyright protection may apply to such images.

The situation may also be slightly different in E.U. countries which apply the sui generis protection for databases. Collections of biodiversity data may be subject to this specific protection against the re-use of a substantial part or the totality of the content of the database. Another exception that researchers should be aware of is the specific photograph protection in some European countries (such as Austria, Denmark, Germany, and Italy). Of course, these specific protection rules apply only in the countries that have introduced them. Outside these countries, the protection has no legal effect.

‘The blue lists’ identify those elements which may reasonably be expected to occur in taxonomic works and, because of their compliance with conventions, lack the creativity that makes copyright applicable. The first list (Patterson et al. 2014) addressed textual components in checklists, classifications, taxonomies, and monographs. Blue2 extends the list with 5 items relating to images. It is the view of the authors that the elements in the list below may be freely re-used unless restricted by a use agreement or a special limitation associated with a few countries. The original source of any re-used element should be cited, but this is demanded by the conventions of scholarship, not by legal obligation. The list may not be complete, and has not been tested in Court.

The first ‘Blue List’ (Patterson et al. 2014) was based on the analysis of the prevailing law and usage patterns, involved a workshop, and input from the community. The analysis led to the conclusion that these elements were not copyrightable. We argue here that the same principle applies to drawings, photos, and maps that illustrate descriptions and circumscriptions of taxa, diagnostic characters, or any other element of the Blue2 list. They do not qualify as copyrightable works as they are executed according to pre-established standards and protocols and are not individual in the sense of copyright.

The situation may differ as far as wildlife illustrations and photos produced during field research are concerned. Such illustrations may be expressed in an individual form and so qualify as works to which copyright may be applied.

Images that do not qualify as copyrightable work and that are not protected by any other intellectual property right, can be reused by any other person for any other legal purpose. Images and collections of images that are protected by copyright or by database protection may only be used with the consent of and under terms set by the rights holder. However, there are situations where even the use of copyrighted material is allowed without authorization.

The rules for these copyright exceptions vary fundamentally in different law systems. While the U.S. applies the so called “fair-use-defense”, European countries aim at the same objective through a catalogue of legally binding rules, called “exceptions and limitations”. In the U.K. and other common-law legislations, the exceptions and limitations are sometimes combined with a “fair-dealing-clause”. The different systems lead to different consequences with respect of the use of copyrighted material.

The “fair-use-clause” is part of the U.S. Copyright Act (Legal Information Institute 1976b: 17 U.S.C. § 107) and means that the unauthorized use of a copyrighted work will not be considered as copyright infringement if this use can be qualified as “fair”. In determining whether there is a fair use, the factors to be considered shall include the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes, the nature of the copyrighted work, the amount and substantiality of the portion used in relation to the copyrighted work as a whole, and the effect of the use upon the potential market for or value of the copyrighted work. The function of the “fair-use-clause” is to give a case-by-case defense to persons who are sued for copyright infringement and where an objective consideration leads to the conclusion that such infringement was done in good faith or does not cause any relevant damage.

The “exceptions and limitations” which are used in the great majority of copyright laws are specific legal rules that authorize uses of copyrighted material for certain purposes. A commonly allowed exception to Copyright law is the use of copyrighted material for research purposes. These rules can be found in the national copyright laws and vary from country to country. The E.U. Directive 2001/29/EC (Eur-Lex 2001) tries to harmonize these rules for the E.U. Member States. It allows, amongst a whole catalogue of other elements, the Member States to provide in their national copyright law for exceptions and limitations for acts of reproduction made by publicly accessible libraries, museums or educational establishments as well as for acts of reproduction or communication for the purpose of illustration for teaching or scientific research. However, as has been illustrated by a recent investigation (Egloff et al. 2014), despite this harmonization effort, national provisions in Europe on copyright and database protection regarding exceptions and limitations for research purposes differ not only in some details but also in substance.

The re-use of copyrighted material even for research purposes may therefore be hampered by current copyright and database protection. The risk is particularly true for international big data studies that were not foreseen by the law-makers and do not fit into the “fair-use”-criteria of U.S. copyright nor will be authorized by any exception rule of European copyright law. Such large projects are likely to inadvertently run counter to some exceptions and limitations or legislation that applies in some national jurisdictions (Uhlir and Clement 2016).

In creative fields, copyright is sometimes justified as a mechanism for encouraging innovation and creativity (Martin 2002). This is based on a very different model than that under which taxonomic researchers typically operate. Producers of creative content are economically incentivized directly by the appeal of their products and their marketability to consumers. Producers of scientific content, particularly in the context of articles for journals, are not incentivized in the same way. Researchers, often with support from public institutions and public or philanthropic grants, typically receive no direct financial incentive to create content. Recent experiments with financial incentivization for creators of scientific content have tended to increase the volume but not the quality of scientific content (Franzoni et al. 2011, Shao and Shen 2011). To the contrary, journals often charge content creators a fee to defray costs associated with page layout, distribution, and other aspects of publishing. Until relatively recently, most journals also sought to acquire all intellectual property rights to the content that they published.

Because taxonomic research is funded in great part by public and philanthropic sources rather than capital investment, it follows that the fruits of this investment and labor are owed to the public rather than to investors. The current practice to cede legal rights to a private publisher, who may use these rights to restrict access to such publications, is highly problematic. The interests of both science and the public are better served by investing in publishing models that maximize accessibility, rather than producing research products paid for by, but not accessible to, the public (Arzberger et al. 2004). Good science depends on independent scrutiny of reported results. When scientific reports survive scrutiny, their value increases. So, lowering access barriers to scientific content provides more opportunity for independent checks and leads to a healthier and more robust science, even when not publicly funded (e.g., citizen scientists). It is also in this context that legal principles concerned with the protection of creative content might not be properly applicable to scientific content.

We do not present our opinion as an indisputable interpretation of the Law, but that it is one of several competing opinions. Laws are not rigid structures but are subject to interpretation. Judicial consensus as to how a law should be interpreted emerges through 'case law' - a success of cases in which the application of the law is contested. We know of no case history that specifically deals with the application of copyright law to biodiversity images.

Competing opinions are held by publishers and by scientists. Until relatively recently, publishers presented themselves as allies of and advocates for scientists, promoting dissemination and awareness of scientific content and further served the scientific community by taking on board the responsibility for shielding scientists from queries about re-use of their materials. In return, they received income from publication fees, sale of publications, and publicity for themselves. Increasingly, the balance has shifted to the commercial model for re-use of scientific literature. This included monopolizing access to their products by, in some cases, seeking transfer of rights from authors to publishers. Authors are put under pressure to assign rights to the publisher to protect the financial interests of the publishers. This has led to a backlash that created many new and exciting new publishing models (Smith et al. 2013).

The historical practices of publishers have established a largely untested opinion as to rights in regard to scientific data, scientific data objects, and publications. Scientists as authors are familiar with this model, some accept it others do not.

Several further arguments have developed from the reliance of scientists on information generated by others. One is the presumption that it is legitimate and reasonable to re-use this information but with the ethical obligation to give credit to those who generated the data. That is, a corner-stone of science is that information is free and openly available.

Another is that research funded through the public purse (as most is) should find its place in the public domain. Taxonomic research is funded in great part by public and philanthropic sources rather than by capital investment. The opinion that the fruits of investment and labor are owed to the public rather than to investors lead to a conclusion that the practice of ceding legal rights to private publishers, who may use these rights to restrict access to such publications, is ethically unsound. Science and the public are better served by investing in publishing models that maximize accessibility, rather than producing research products paid for by, but not accessible to, the public (Arzberger et al. 2004).

A final opinion is the value of subjecting reports to scrutiny. This process can correct errors, show weaknesses in arguments, point to other publications that are not cited. That is, scrutiny adds value. The traditional publishing model adds independent scrutiny by two or three reviewers of manuscripts. Other models that make content freely accessible, even before publishing, provide increased opportunity for independent checks and leads to a healthier and more robust science. The malleable nature of scientific products further undermines any case that copyright principles to apply to scientific content.

The value of the competing opinions regarding copyright law as it applies to images in the biological sciences has not been tested in Court. There is as yet no judicial consensus. We argue that the U.S. Supreme Court decision in Feist Publications, Inc. v. Rural Telephone Service Co. (Legal Information Institute 1991: 499 U. S. 340), in which the re-use of telephone numbers was not seen as an infringement of copyright is relevant. Similarly, rulings against the copyrightability of non-creative photographs (Reinbothe and von Lewinski 2002, Swiss Federal Court 2004) or against the protection of technical manuals (Oberlandesgericht Düsseldorf 2002, Swiss Federal Court 2008) are relevant to the re-use of content in a standard format).

The principles of scholarship in taxonomic research include the expectation that relevant previous work be cited. Citation of publications identifies prior work and helps to assure reproducibility and comparability of the results of scientific research.

Citation offers a mechanism of providing credit to work by others, that is, attribution. In an increasingly digital world, we should be attentive to the principles of citation, comply with any legal obligations, and identify those who acquired the data or in any way contributed to the supply chain and/or added value to data (Patterson et al. 2010).