Plant taxonomists are responsible for, among many other tasks, the formalisation of new taxa, which require the publication of new names for them. According to the current version of the International Code of Nomenclature for algae, fungi, and plants (Turland & al., 2018—hereafter ICN), either a description or a diagnosis is enough for the valid publication of a name. In the cases that neither is provided, a name is not validly published (Art. 38 of the ICN), and is commonly referred to as a nomen nudum (“naked name”).

In fact, Art. 38.2 of the ICN establishes that “A diagnosis of a taxon is a statement of that which in the opinion of its author distinguishes the taxon from other taxa”, while in the ICN glossary, a description is defined as: “a published statement of a feature or features of an individual taxon; a description (or a diagnosis) is required for valid publication of a name of a new taxon (Art. 38.1(a) and 38.3); a validating description need not be diagnostic”. The current version of the ICN is clear regarding the distinction between diagnosis and description, so that the definition of both terms does not currently seem to be a problem. The same approach is confirmed in Turland (2019: 18). It should be remarked that the ICN is focused on nomenclature and not on taxonomy, and does not aim to judge whether descriptions and diagnoses adequately represent the taxa (Nicolson, 1991). Furthermore, we highlight that the discussion presented here refers to Plant Taxonomy, not to other groups of organisms also covered by the ICN, i.e., algae and fungi.

From a historical point of view, Linnaeus (1751), in his Philosophia botanica, gave the definition of a descriptio in the Adumbratio 326 (p. 256) as follows: “Descriptio […] est totius plantae character naturalis, qui describat omnes ejusdem partes externas”, and then he gave more details on how to set up and improve a description: for Linnaeus, a descriptio is an analytic statement clearly and conceptually distinct from a diagnosis, which is a synthetic statement. More recently, Ghiselin (1997), in the glossary at the end of his book, stated that a description “enumerates the properties of things, irrespective of whether or not the properties in question are defining” and a diagnosis “enumerates properties that are useful in identification”, thus highlighting the descriptive aspect of a description, which aims at completeness, and the comparative aspect of a diagnosis, which aims at succinctness. Furthermore, a diagnosis reflects the “type method” that represents the epistemological point of contact between Taxonomy and Nomenclature (Candolle, 1867; Mayr, 1989; Witteveen, 2015, 2017, 2018). More reflections on this topic can be found in Simpson (1961), Wiley (1981) and Winston (1999).

Despite the explicit and satisfactory differentiation in the ICN, we argue that the distinction of a diagnosis and a description is not clear to many taxonomists these days, especially the younger ones. New taxa, especially new species, are often described supported only by descriptions, without a diagnosis (e.g., Berry & Galdames, 2013; Van der Maesen, 2013; Palchetti & al., 2018; Shui & al., 2019; Vaezi & al., 2019; Vladimirov & al., 2019), or other times diagnostic and descriptive information is joined under one or the other (e.g., Kuijt & Delprete, 2019). In some cases, a diagnosis is presented after a description (e.g., Arigela & al., 2019; Guzmán‐Guzmán, 2019; Xiao & al., 2019), which we consider that further adds to the current state of confusion. Considering the fundamentally distinct purposes of diagnoses and descriptions (see above), we argue that it would be for the benefit of Plant Taxonomy, taxonomists and users of taxonomic classifications if both a diagnosis and a description were always provided to formalise new taxa and that, for consistency, diagnoses be presented before descriptions for each taxon.

Nevertheless, the importance of distinguishing diagnoses and descriptions goes much beyond the formalisation of new taxa. In fact, monographs and other taxonomic literature presenting morphological information should ideally present both diagnoses and descriptions for taxa. This would maximise the usefulness of those treatments, in allowing distinguishing a taxon from its relatives (e.g., a species from its congeners) in the most succinct manner, which is achieved by means of a diagnosis, and also in informing characters of the taxon in question as thoroughly as possible, which is achieved by means of a description. Synoptic works, which normally do not present descriptions of taxa, could nevertheless provide diagnoses for the taxa treated—those diagnoses, although succinct, would have enormous usefulness for the readership in order to comprehend the species concepts and delimitations adopted by the author. Currently, the vast majority of taxonomic works being published do not provide diagnoses for taxa that are not being newly described, a situation that we hope to change with the present letter.

Traditionally, the characters used for descriptions are morphologic, but with the development of new technologies, other types of information could be used, such as, e.g., chromosome number and morphology, physiological characters, biochemical characters, and DNA molecular data (e.g., Goldstein & DeSalle, 2011; Jörger & Schrödl, 2013; Renner, 2016; Bakker, 2017; Viruel & al., 2019). It is undeniable that non‐morphological information can be very useful for supporting more stable and refined taxonomic classifications (Jörger & Schrödl, 2013), offering important support to morphology (although sometimes contradicting it), and most probably will see crescent use among systematists. The use of these extra types of information is undoubtedly improving the informational content for Taxonomy and Systematics as a whole. Such integrative approaches are critically important especially for the study of species complexes and cryptic species, and constitute further support for the integration (but not substitution!) of non‐morphological information to the elaboration of descriptions (Tripp & Lendemer, 2014). We acknowledge that information on micro‐ or nanomorphological features such as chromosomic and molecular data is not always available, but its inclusion in a description is desirable and should be done when possible.

As an illustrative example, Li & al. (2012) recently published a new fern genus, Gaga Pryer & al., presenting a description and mentioning, regarding the etymology of the new taxon, that “At nucleotide positions 598–601 in the matK gene alignment, all Gaga species have ‘GAGA’ […], a sequence pattern not seen at this site in any other cheilanthoid fern sampled”, from which the name of the genus was dedicated to a famous American pop star. Li & al. (2012) were the first to use a nucleotide sequence from which they justified the etymology of a new genus name, but they omitted this important molecular information from the description they provided for the new genus. This very relevant molecular information could have been included in the description of the new taxon, instead in the Etymology section. Furthermore, it should be noted that the first paragraph of the description they provided for Gaga is clearly a diagnosis, which is, however, not referred to as such; this fact corroborates the prevailing view, which we highlight here, that diagnoses and descriptions are nowadays being confused by many among the scientific community.

In the case of diagnoses, however, we argue that the use of non‐morphological information would undo their very purpose, i.e., to provide the most succinct and accessible means for the identification of the taxon in question. Therefore, we argue that diagnoses should use only morphological characters. The use of morphological diagnoses advocated here does not preclude that non‐morphological characters be used to elaborate non‐morphological diagnoses, e.g., a molecular diagnosis presenting a string of nucleotides that is unique to the taxon in question. Thus, contrary to the description, the combination of different types of characters is counterproductive for diagnoses.

The abandonment of the use of morphology for the description of new taxa and for the taxonomic classification as a whole has been suggested in some recent works (e.g., Cook & al., 2010). We feel that this would have extremely negative consequences to Taxonomy and consequently to Systematics, because most of the taxonomic novelties (especially in Plant Taxonomy) are happening in contexts where molecular works are completely unavailable. Furthermore, people working with molecular phylogeny often lack experience and knowledge of taxonomic practices and nomenclature, and there is a well‐known general trend of reduction (even extinction, in some environments) of taxonomists in research institutes, universities and even museums (Agnarsson & Kuntner, 2007; Ebach & al., 2011; Wägele & al., 2011; Sluys, 2013). The development of new techniques is increasing, not decreasing the demand for taxonomic expertise and correct specimen determinations (Will & Rubinoff, 2004; Packer & al., 2009; Taylor & Harris, 2012). In sum, abandoning morphology would bring no benefits to Plant Taxonomy and would effectively stall taxonomic advancement in the regions of the world precisely where most of the unknown biodiversity occurs. This would also have nefarious consequences for biodiversity conservation, not only because many narrowly endemic species would remain unknown to science, but also because without the use of morphology, it would become essentially impossible to recognise or determine rare and/or threatened species (Ely & al., 2017; Thomson & al., 2018).

To conclude, we think that a more disciplined use of diagnoses and descriptions should be adopted in Plant Taxonomy. We think that it is most beneficial for Systematics and its users that both a diagnosis and a description be informed in taxonomic works, both for the description of new species and for the treatment of already described taxa. We think that synoptic taxonomic works should inform the diagnostic characters for the taxa treated. Furthermore, we argue that diagnoses and descriptions should not be joined, or considered equivalent or interchangeable. In order for these changes to be made possible, it would be necessary that the editors of journals publishing taxonomic literature demand from authors the proper use of diagnoses and descriptions. To sum up, we present below the concepts we adopted for a diagnosis and a description.

A diagnosis is a synthetic statement of the morphological characters that allow the distinguishing of the taxon in question from its relatives (e.g., a species from its congeners). A diagnosis should ideally be as concise as possible.

A description is an analytic statement describing features that characterise the taxon in question, including macro‐morphological to anatomical, biochemical, karyological and molecular aspects. A description should ideally be as thorough as possible.

植物分类学家负责许多其他任务，包括使新的分类单元正式化，这需要为其发布新名称。根据当前版本的《藻类，真菌和植物国际命名法》（Turland等人，2018年，以下简称ICN），有效地使用名称的描述或诊断就足够了。在未提供任何名称的情况下，名称不会被有效发布（ICN第38条），通常被称为nomen nudum（“裸名”）。

其实是艺术。ICN的第38.2条规定：“分类单元的诊断是其作者认为该分类单元与其他分类单元相区分的说明”，而在ICN词汇表中，描述被定义为：“单个分类单元的一个或多个特征；有效发布新分类单元名称需要描述（或诊断）（第38.1（a）和38.3条）；验证描述不必是诊断性的。” 关于诊断和描述之间的区别，ICN的当前版本很明确，因此两个术语的定义目前似乎不是问题。Turland（2019：18）也证实了相同的方法。应该指出的是ICN专注于术语而非分类法，并且不旨在判断描述和诊断是否足以代表分类单元（Nicolson，1991）。此外，我们强调指出，此处提出的讨论是针对植物分类学，而不是ICN也涵盖的其他生物体，即藻类和真菌。

从历史的角度来看，林奈（1751）在他的《植物哲学》中对Adumbratio 326（p。256）中的描述进行了如下定义：“描述植物的自然特征，描述人类的行为。 ”，然后他提供了更多有关如何建立和改进描述的细节：对于Linnaeus而言，描述是一种分析性陈述，在概念上与诊断有明显区别，后者是一种综合性陈述。最近，Ghiselin（1997）在其书末的词汇表中指出，“对对象的属性进行枚举，而不管所讨论的属性是否在定义中”；而对诊断的描述则“对可用于识别的属性进行枚举”，因此强调了以完整性为目的的描述的描述性方面，以简洁性为特征的诊断的比较性方面。此外，诊断反映了“类型方法”，它代表了分类学和命名法之间的认识论联系（Candolle，1867年； Mayr，1989年； Witteveen，2015年，2017年，2018年）。关于这一主题的更多思考可以在辛普森（Simpson，1961年），威利（Wiley，1981年）和温斯顿（1999年）。

尽管在ICN中有明显且令人满意的区分，但我们认为，对于当今的许多分类学家，尤其是较年轻的分类学家来说，诊断和描述的区分尚不清楚。新的类群，特别是新物种，通常仅以描述来支持，而没有诊断（例如Berry＆Galdames，2013年； Van der Maesen，2013年； Palchetti等人，2018年； Shui等人，2019年； Vaezi等人， 。，2019 ; Vladimirov等人，2019），或者将诊断和描述性信息结合在一起（例如Kuijt＆Delprete，2019））。在某些情况下，会在描述后给出诊断（例如Arigela等，2019；Guzmán-Guzmán，2019； Xiao等，2019），我们认为这进一步加剧了当前的混乱状态。考虑到诊断和描述的根本不同目的（请参见上文），我们认为，如果始终提供诊断和描述来使新的分类单元正式化，这将有利于植物分类学，分类学家和分类学分类的使用者，为了保持一致性，在对每个分类单元进行描述之前先进行诊断。

然而，区分诊断和描述的重要性远远超出了新分类单元的形式化。实际上，提供形态学信息的专着和其他分类学文献理想地应提供分类学的诊断和描述。这将最大程度地利用这些治疗方法，从而可以以最简洁的方式将分类单元与其亲属（例如物种与其同类动物）区分开来，这可以通过诊断来实现，也可以告知分类单元的特征。尽可能彻底地解决问题，这是通过描述来解决的。通常不会提供分类单元描述的概要工作仍然可以为所治疗的分类单元提供诊断-尽管简明扼要，对于理解作者所采用的物种概念和划界方法将具有极大的帮助。当前，正在出版的绝大多数分类学著作都没有提供对未被新描述的分类学的诊断，这种情况我们希望随本信而改变。

传统上，用于描述的字符是形态的，但是随着新技术的发展，可以使用其他类型的信息，例如染色体数目和形态，生理特征，生化特征和DNA分子数据（例如Goldstein） ＆DeSalle，2011 ;Jörger＆Schrödl，2013 ; Renner，2016 ; Bakker，2017 ; Viruel等，2019）。不可否认，非形态学信息对于支持更稳定和完善的分类学分类非常有用（Jörger＆Schrödl，2013），为形态学提供了重要的支持（尽管有时会相互矛盾），并且很有可能会在系统论者中看到新月形用法。这些额外类型的信息的使用无疑会改善分类法和系统学的信息内容。这种整合方法至关重要，特别是对于物种复合体和隐性物种的研究而言，并为非形态学信息的整合（但不是替代！）提供了进一步的支持，以用于描述的详细说明（Tripp＆Lendemer，2014）。我们承认，关于微观或纳米形态特征（例如染色体和分子数据）的信息并不总是可用，但是将其包含在描述中是可取的，应在可能的情况下进行。

作为说明性示例，Li等人。（2012）最近发表了一个新的蕨类属，Gaga Pryer等，对新分类单元的词源进行了描述和提及，“在matK基因比对中的598-601位核苷酸处，所有Gaga物种都有'GAGA “ […]，这是在任何其他采样的兰藻类蕨类植物中都没有发现的序列模式”，该属的名称专门用于美国著名的流行歌星。Li＆al。（2012年）是第一个使用核苷酸序列证明其新词源的词源合理的核苷酸序列的人，但他们从为该新物种提供的描述中省略了这一重要的分子信息。此非常相关的分子信息可能已包含在新分类单元的描述中，而不是在“词源”部分中。此外，应该注意的是，他们为Gaga提供的说明的第一段显然是诊断，但是并未如此。这一事实佐证了一种普遍的观点，我们在这里强调这一点：如今，诊断和描述已被许多科学界所混淆。

但是，在诊断的情况下，我们认为使用非形态学信息将无法达到其目的，即提供最简洁，可访问的手段来识别相关的分类单元。因此，我们认为诊断应仅使用形态特征。此处提倡使用形态学诊断并不排除使用非形态学特征来阐述非形态学诊断，例如，分子诊断显示出所讨论的分类单元所特有的一串核苷酸。因此，与描述相反，不同类型字符的组合对诊断起反作用。

在最近的一些工作中已经提出放弃使用形态学来描述新的分类群和将分类学作为一个整体进行分类（例如Cook和al。，2010）。我们认为这将对分类学以及因此对系统学产生极大的负面影响，因为大多数分类学新颖性（尤其是在植物分类学中）都是在分子工作完全不可用的情况下发生的。此外，从事分子系统发育研究的人们通常缺乏分类学实践和术语的经验和知识，研究机构，大学甚至博物馆中分类学家的减少（在某些环境中甚至灭绝）是众所周知的普遍趋势（Agnarsson＆昆特纳，2007年；埃巴赫等人，2011年; Wägele等人，2011年；Sluys，2013年）。新技术的发展正在增加，但并未减少对分类学专业知识和正确的标本测定的需求（Will和Rubinoff，2004年； Packer等人，2009年； Taylor＆Harris，2012年））。总而言之，放弃形态将不会给植物分类学带来任何好处，也不会有效阻止世界上大多数未知生物多样性发生的确切地区的分类学进展。这不仅会给生物多样性的保护带来灾难性的后果，不仅因为许多狭end的特有物种在科学上仍然是未知的，而且还因为不使用形态学，就几乎不可能识别或确定稀有和/或受威胁物种（E​​ly和等人，2017年； Thomson等人，2018年）。

总而言之，我们认为在植物分类学中应该对诊断和描述进行更严格的使用。我们认为，对系统学及其用户来说，诊断和在分类工作中应提供描述，既包括对新物种的描述，也包括对已经描述的分类单元的处理。我们认为，天气分类学工作应告知所治疗分类单元的诊断特征。此外，我们认为诊断和描述不应合并，也不应视为等同或可互换。为了使这些更改成为可能，出版分类学文献的期刊编辑必须要求作者正确使用诊断和描述。综上所述，我们在下面介绍用于诊断和描述的概念。

甲诊断是形态特征，其允许该分类单元的显着从其亲属（例如，从它的同族种）问题的合成语句。理想情况下，诊断应尽可能简洁。

甲描述是描述所讨论表征分类单元的功能，包括宏观形态解剖，生物化学，核型和分子方面的分析语句。理想情况下，说明应尽可能详尽。