Aurivillius’s “Neue oder wenig bekannte ColeopteraLongicornia” (1886–1927), the correct years and page numbers

Abstract Aurivillius’s work entitled “Neue oder wenig bekannte ColeopteraLongicornia” was published in parts over a period of over four decades. There were two page numbers on most pages of these publications, one ordered by Aurivillius, the other by the journal. Historically, different authors have used different page numbers, and sometimes different years for these publications, which has caused chaos in the citations. Herein, accurate dates of publications for this work, and correct page numbers that should be used are provided and discussed.


Introduction
Christopher Aurivillius (1853Aurivillius ( -1928 was a very important Swedish entomologist, who published 67 references regarding Cerambycidae from 1886 to 1929 (Tavakilian and Chevillotte 2019). Among them, 20 parts were titled as "Neue oder wenig bekannte Coleoptera Longicornia" and numbered from 4 to 23. Most of them (except the 8 th part) have two page numbers printed on each page, both of which have been cited by many different authors. In order to determine the correct page numbers and the accurate dates of publication for this significant work, we analyzed all the Cerambycidae literature of Aurivillius.

Methods of literature collecting
We accessed literature in three ways for this study: a) downloaded pdf files from the Biodiversity Heritage Library: http://www.biodiversitylibrary.org/; b) copied the original pages directly from library holdings (the first author visited the libraries of the Institute of Zoology, Chinese Academy of Sciences, Beijing, China; National Science Library, Chinese Academy of Science, Beijing, China; Muséum national d'Histoire naturelle, Paris, France; Division of Plant Industry, Florida State Collection of Arthropods, Gainesville, Florida, USA; and the National Museum of Natural History (Smithsonian Institution), Washington DC, USA, etc.); c) solicited help from colleagues (especially G. Tavakilian and S. Lingafelter).

Dating the publications
In researching the dates of publication for this work, we consulted five points of reference: a) date printed on first and last pages; b) date shown by the Zoological Record; c) date used by Aurivillius's catalogues; d) date used by literature citing related references; e) date printed on original wrapper.

Common errors encountered when citing this series of papers
The errors occurred in the date (see Table 1), page numbers (see Table 1), information regarding the figures and plates, part numbers, first and last page numbers, journal volume numbers, and so on. a) Errors regarding the separate plates. There were two kinds of figures in this work, text-figures were inside the content and provided with continuous numbers (see Table 1), while end-plates were printed as separate plates, normally numbered from one. The former can be ignored in the reference, while the latter should be added. For example, Hüdepohl and Heffern (2004) wrote the reference as " Aurivillius, C. 1907. Neue oder wenig bekannte Coleoptera Longicornia. 9. Arkiv för zoologi.  Aurivillius's page number did not have page 58. That was because every new part began with odd numbers. Therefore, when the previous part ended with odd numbers (parts 6, 9, 12, 13, 15, 16, 17, 19, 20, 22), one even number would be taken by an empty page (page numbers 58,228,264,334,404,480,502)  Aurivillius chose the journal AFZ since the eighth part and did not change it again. In this first printing in the AFZ journal, the journal did not include Aurivillius's own page numbers. But from the ninth part, the AFZ printed both the journal's page number on top and Aurivillius' own page number at bottom. The date of publication of the eighth part is confusing. The printing date on the last page is "Tryckt den 27 november 1903", so it should be 1903. However, Aurivillius (1912;  The correct publication date of the ninth part is surely 1907, and most of authors cited it correctly (Aurivillius 1912Breuning 1944;Gressitt 1951;Podaný 1971;Nakamura et al. 1992;Hüdepohl and Heffern 2004;Heffern 2005Heffern , 2011 The publication date of the 11 th part is confusing. The correct year should be 1910, since on the last page it was printed "Tryckt den 24 september 1910", Aurivillius himself cited it as 1910 (Aurivillius 1912, and the Zoological Record also listed it as 1910. Most authors cited it correctly (Quentin 1956;Podaný 1968;Hüdepohl 1992;Martins 1997Martins , 1998Martins , 2002Vives and Heffern 2001;Heffern, 2002;Vives and Abang 2003;Monné 2005aMonné , 2012Monné , 2019aBousquet et al. 2009;Bentanachs et al. 2010  , and the Zoological Record also included it in 1914. Most authors cited it correctly (Adlbauer 2002;Juhel and Bentanachs 2011;Juhel 2014a The species number 523 was used twice. It was used for "523. Hilarolea humeralis" in the 16 th part, and again as "523. Ophistomis splendida" in the 17 th part. The publication date of this part is surely 1920 and few errors were made. Martins (2005) carelessly used 1922 but he corrected it to 1920 later (Martins 2011(Martins , 2014 The publishing year of the 18 th part is surely 1922. However, some authors used the submission date 1921 (e.g., Makihara, 1999), or even 1920 for unknown reasons (e.g. Makihara and Woro 2002).
Publishing The date of publication of the 19 th part is one of the most confusing cases, since different people used different years and for different reasons.

Date of publication we chose
The dates of publication of this series of work contain several confusing cases; the detailed information is shown in Table 1. For parts 5 and 6, we chose the earlier date indicated by the Zoological Record and Derksen and Scheiding (1963) When we talk about "distribute reprints in advance" in Aurivillius's cases, the authors mean distribute the reprints after the printing date ("tryckt den XX YY 19ZZ") but before the distribute date of the publisher (either printed on the wrapper, normally for the whole volume, or date applied subsequently by the Zoological Record).

Why the journal page numbers should be used
For parts 8 to 23 of Aurivillius's works, the reasons that the journal page numbers should be used include: 1) the works were first officially published in the journal; 2) the large book titled "Neue oder wenig bekannte Coleoptera Longicornia" does not exist; 3) Aurivillius himself used the journal page numbers instead of his own page numbers (Aurivillius 1912(Aurivillius , 1928; 4) if Aurivillius's own page numbers were chosen, the results are chaotic since the numbers continued between different journals, different years, and additionally, some parts were missing (Table 1: pages 1-14 and 77-92 were not printed); 5) if Aurivillius's own page numbers were chosen, logically there should be pages preceeding them in the same volume. For example, considering "Arkiv för zoologi 13(9): 361-403" instead of "Arkiv för zoologi 13(9): 1-43", logically there should exist "Arkiv för zoologi 13(9): 1-360" (or "Arkiv för zoologi 13: 1-360"), but this is not the case.
How to identify which page number was the journal's page number 1) the page number was printed on the upper left corner (of even pages) or the upper right corner (of odd pages), which was the style of the journal "Arkiv för zoologi" (Aurivillius 1917(Aurivillius , 1919(Aurivillius , 1926, except the first page normally appeared on the lower right corner; 2) each part of each volume was numbered from one, which was also the style of journal "Arkiv för zoologi" at that time (Aurivillius 1917(Aurivillius , 1919(Aurivillius , 1926).

Aurivillius's own numbers might be chosen for the following reasons
1) it was the choice of the Titan database (Tavakilian and Chevillotte 2019), which is the most exhaustive Cerambycidae database; 2) larger sized numbers appear more important (for some reasons), for parts 3 to 7, which also had two page numbers printed, all were cited with the correct journal's page numbers, because they are larger than Aurivillius's own page numbers (such as Wappes et al. 2011;Ślipiński and Escalona 2016;Souza 2016;Tavakilian and Chevillotte 2019); 3) page numbers on the mid-bottom are more noticeable than page numbers on upper left corner (of even pages) or upper right corner (of odd pages); 4) works were reprinted with the smaller page numbers even though they were originally from a book or journal with the larger page numbers; realizing this subsequent workers may have chosen the larger numbers; 5) to follow author's citing Aurivillius's own page numbers.

The trend
From Fig. 5 we can see that more than half of authors used Aurivillius's page numbers instead of the journal's page numbers. However, from Fig. 6 we can see that more authors used the journal's page numbers than Aurivillius's page numbers before the year 2000, while most authors used Aurivillius's page numbers after the year 2000. Analyzing the references in more detail (Fig. 7), we can see that all authors before 1990 used the journal's page numbers, while more and more authors used Aurivillius's page numbers after 1991. The reasons for this trend might include: a) young authors did not know the history and might choose the bottom page numbers by the first glance; b) many current authors use the Titan database and copy the information from the website.   We hope that the Titan database will correct the information and use the journal's page numbers after reading this paper, and authors in the future will cite the related references in correct way.

Correct citation of Aurivillius's works
Based on the above analyses, we suggest that in the future authors cite the work of Aurivillius as follows: the journal page number must be included, Aurivillius's page numbers might be included inside square brackets [] or not included, the internal figure numbers (text-figures) can be included or not, while the supplemental information for the end-plates must be included. Aurivillius, C (1927b)