In the research of scientific and mathematical history, efficient skills in searching for and accessing old academic journals (preferably for free) are essential. However, even when using platforms like Internet Archive or Google Books, it's often difficult for me to find what I'm looking for, and even if I do find it, full-text access might not be available. Then, when I ask for help on this forum, someone always manages to find what I need. Is there some trick I'm unaware of?

For instance, right now I'm trying to find a pioneering paper from 1926 published in Germany titled "Winkelvariable und Kanonische Transformationen in der Undulationsmechanik," which applies Hilbert space theory to quantum mechanics. The bibliographic information is Zs. f. Phys., 40, 193-210 (1926), but I'm having trouble locating it.

By the way, I recently asked for help finding another paper by the same author in the preceding issue of the journal on this forum, and several people found it for me. Where does this difference in skill come from? Please, I'd like to know the trick.

  • 2
    $\begingroup$ The New York Public Library's reference service, AskNYPL, is famed for the skills of its personnel in tracking down difficult or obscure volumes. While not a definitive resource, they are a useful tool to know and leverage. May require an NYPL id but as a public institution, the barriers to obtaining one are very low. nypl.org/ask-nypl/about $\endgroup$
    – DJohnson
    Commented May 22 at 0:50
  • 3
    $\begingroup$ Others might not be using free services to find resources. That doesn't mean they are paying directly for those services. They could be affiliated with institutions that pay for those resources (e.g., library databases), and those institutions may be providing access to those resources to employees, faculty, students, etc. $\endgroup$ Commented May 22 at 4:53
  • 1
    $\begingroup$ The journal Zeitschrift für Physik is still available online, and I think the paper you're looking for is doi.org/10.1007/BF01400361 (from 1927). I guess the issue is that you don't have the subscription access? $\endgroup$
    – Anyon
    Commented May 24 at 4:44
  • $\begingroup$ I don’t, Anyon. $\endgroup$
    – enjin2000
    Commented May 24 at 8:05

5 Answers 5


For Google Books you will often want to use date-restricted searches (for some reason these don't work so well after the early 1900s, at least for me), but don't restrict to a single publication year because (especially in the case of 1800s literature) these can vary by $2$ or $3$ years according to how the year was determined (by conference or presentation meeting date, by journal issue date, by journal volume date, etc.). Also, phrase searches for the paper's title are usually best, but sometimes you'll have to remove the quotes used for a phrase search because the .pdf-to-text conversion for searches can mess up some words. Indeed, sometimes it's best to leave out a few words in the title, especially when the original publication is written in old-style English or gothic German style (something you usually don't know in advance, so you do this later on when more obvious searches are not working). Sometimes it's best to include the author as an additional search word (but use only LAST/FAMILY name), and sometimes it's best to not include the author. Moreover, Google searches (regular and Book) are sometimes random-like and inconsistent with results given, especially for obscure stuff (no hits or only 2 or 3 hits), such as using MORE search words/phrases can sometimes produce ADDITIONAL hits, so you'll want to be very persistent in searches and not try to use logical deduction when trying different searches (e.g. thinking that adding a word cannot increase search results; substituting what should be an equivalent form, such as the German word über/uber/ueber, will not change the search results).

Besides Google Books $\ldots$

Numdam -- Has.pdf files of individual papers can be downloaded, pretty much everything from French 1800s math journals. Some of the links I give in the following answer are to Numdam URLs: Reference Request - Functions of the form $\frac{ax^2+bx+c}{dx^2+ex+f}$. Also, although I think all (or nearly all) of the papers cited in the following answers are to Google Books, probably all of the French journals mentioned are at Numdam: Evaluate $\lim_{x\to 0}\frac{x-\sin x}{x\sin x}$ without to use L'Hopital AND Conditionally convergent series AND What set of criteria led Hamilton to discover the quaternions? (in this answer the journals are where the reviews are published) AND differential equation of conics.

HathiTrust -- A huge number of old books and old journals can be found here, most of which I can download page-by-page, but entire document download (by journal article, or by book/journal volume) requires affiliation with a member institution, which I don't have.

Biodiversity Heritage Library -- Has a large number of old scientific society journals, which are where a lot of math papers before the last few decades of the 1800s were published. From here I've downloaded complete sequences up to the mid 1920s of many journal/periodical volumes, such as Report of the British Association for the Advancement of Science and similar U.S. scientific associations. See my comments to Around 1904, did the scientific community take the atomic hypothesis seriously? AND this answer to What is the history of linear vs logarithmic scales?

Royal Society of London Catalogue of Scientific Papers 1800-1900. Volume I, Pure Mathematics (1908) -- This is an essentially complete list of papers, categorized by topic/subject, fitting the restrictions of its title. English translations of the paper titles are used (often shortened when the original title is long), which makes searching for the papers difficult (can't use phrase or word searches in online resources) but there is the advantage that the translated titles somewhat indicate what the paper deals with. I've found this so useful that I obtained a print-on-demand hardcopy around 12-14 years ago.

Best way to find papers in Mathematics by a specific author? -- My answer to this Mathematics Stack Exchange question lists many useful sites for math literature searching.

Educational Times and Mathematical Questions and Solutions from the Educational Times -- I'm listing this periodical individually only because I discussed which volumes appear to exist online in my answer to What is "educ times"? A journal? in Academia Stack Exchange, and thus it was probably not seen by many participants in this Stack Exchange who might be interested in this information.

Google search tips for information about authors -- See my answers to Who is Donald L. Webb AND Who was N.M. Stephens who refuted the Stronger Feit-Thompson Conjecture?


A famous pirate site with a long legal history is Library Genesis (on Wikipedia you can read about its ups and downs). A trick that someone (not me, of course) might have used is to retrieve the article on Library Genesis, copy distinctive sentences from the article and then search those sentences on Google to see if there is a publicly available version somewhere.

That said, I'll also give a list of my favorite libraries:

Let's spend a few words on HathiTrust, which is perhaps the richest library: the resource is often available, but only for the United States; to overcome the problem of not residing physically in the USA, you can use a VPN if it's legal in your country. For example, in my country (Italy) it's legal, but it's also legal in Japan and in almost every country in the world. The easiest way to get a VPN is to install the Opera browser and set the VPN to Americas (Americas may also mean Brazil, but it seems I am lucky because whenever I needed HathiTrust, I always got an IP from the USA).

To get London's article, I went to HathiTrust with Opera, searched for Zeitschrift für Physik (sometimes it gives more than one result because the same journal might have been digitized in multiple universities) and then went to volume 40. London's article is here.

  • $\begingroup$ It says “This item is not available online.” $\endgroup$
    – enjin2000
    Commented May 24 at 8:11
  • 1
    $\begingroup$ @enjin2000 were you by any chance using a VPN to simulate being in the US? $\endgroup$
    – mdewey
    Commented May 24 at 14:56
  • 1
    $\begingroup$ No way, mdewey. Actually, I wasn't familiar with the term "VPN" at all until just now. $\endgroup$
    – enjin2000
    Commented May 24 at 15:18

Depending on how far back in time you want to go, have a look at this page I put together a couple years ago: Navigating Historical Learned Societies

This covers a couple 1600s-1700s royal societies and can serve as a starting point of traps to watch out for. Once you hit the 1800s things can get a little harder to follow because of the larger number of journals and such that came into existence. By the time you get to the early 20th century you have to contend with a lot of abbreviations that are not always clear, and they abbreviate other languages (notably German) that might be difficult to parse until you know the basics of that language. I have a notes page that I jot things down into, which can also help get someone started:

Journals notable for quantum physics:

  • Sitzungsberichte Der Königlich Preussischen Akademie Der Wissenschaften Zu Berlin
  • Annalen Der Physik
  • Monatsberichte der Königlichen Preussische Akademie des Wissenschaften zu Berlin
  • Sitzungsberichte der Preussischen Akademie der Wissenschaften
  • Verh. Dtsch. phys. Ges. = Verhandlungen der Deutschen Physikalischen Geselleschaft (Verhandlungen der DPG)
  • Ann. Physik. = Annalen der Physik
  • Séances Soc. Fr. de Phys. = Séances de la Société française de physique
    • Available on Google Books, three years available from Gallica.bnf

Some German terminology that might help:

  • Sitzungsberichte, Reports
  • Verhandlungen, Transactions
  • Mitteilungen, Communications or Notes
  • Zeitschrift, Magazine or Journal
  • Naturwissenschaften, Natural Sciences
  • Monatsberichte, Monthly Reports

And generally searching something like "abbreviations Verh. Dtsch. phys. Ges." will turn up many links to lists of journal abbreviations.


You may look at my web page: https://www.math.purdue.edu/~eremenko/ where I collected some resources, for mathematics.

If you have an account in some university library, you can always use the Interlibrary Loan (ILL) service, which will usually deliver you a pdf file of any paper within few days. To request a paper you need a precise reference, and this can be usually found in Zentralblatt Math (which is free).

And of course, you can use the general tools, Google and Google Scholar. Finally there are "pirate" sites to which I give no links since a) these links are frequently changing, and b) their legal status is unclear (may depend of the country).

Research in history may require older papers which are not covered by Zentralblatt. For this, there is a good source:

Poggendorff, Biographisch-literarisches Handwörterbuch zur Geschichte der exakten Wissenschaften, Leipzig, 1863-1904.

originally in 4 volumes.

It is supposed to cover ALL publications in mathematics and natural sciences before the beginning of 19 century, with short biographies of the authors. This edition was continued until 1960, but the first 4 volumes are especially useful since they cover what is not covered by other sources. This Wikipedia article contains the links to scanned copies. Unfortunately this never was decently digitalized.


When researching answer for this site, I frequently make use the online collection Gallica of the Bibliothèque nationale de France when I look for French publications in mathematics and the natural sciences. Göt­tinger Digi­tali­sierungs­zentrum is my first stop whenever I am looking for German publications in mathematics. For occasional searches of Dutch sources I have made use of the Delpher service of the Royal Dutch Library.

The interfaces of freely accessible national online collections are not necessarily focused on international users and basic knowledge of the national language tends to be quite useful when conducting searches. Even when there is no language barrier, searching may be cumbersome, the typical problem being too few or too many hits.

A significant problem in finding specific publications that I often encounter with historical references is resolving shortened or translated journal names into their proper name. Example: Crelle or Crelle's Journal is actually named Journal für die reine und angewandte Mathematik, whose founding editor was August Crelle.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.