This text of the Shatapatha, Maadhyandina Shaakha, was encoded in 1971, partly funded by the National Endowment for the Humanities, by Dr. H. S. Ananthanarayana under the direction of Dr. W. P. Lehmann. It is painstakingly reconstructed and accessible in a multitude of ways.
Revised site note January 24, 2010: I made a tribute page to their work which I preserve as-is as part of the palimpsest of this site, you can view that page here. And a note on what I mean by palimpsest can be found here.
It has been extensively re-worked, as its original form reflected the early limits of technology at that time which were, nonetheless, transcended by the vision of the project directors. NOTE: the word breaks per akSara are not present. This is for the reasons of
Also, long vowels are rendered as double letters, with accent via ' markings over the appropriate letter. Font constraints require that vocalic R remain a capital, with : for length, and the original ! for accent.
The ZB normally is considered to mark only udaatta in its mss (cf. Weber's edition). There are, in fact, two markings as you will find in the text. A traditional stroke ' over a vowel as well as "&" following it. The Texas team preserved both. The issue at hand involves how to read them. Contrary to Weber and Caland, however, there have been several arguments to the contrary. Specifically from Hoffman, Kiparsky, Witzel, and Cardona--see discussion. In essence, the traditional ' stroke (an underline in the mss) means NOT udaatta, but instead anudaatta. This is a thorny issue stretching into matters of various intonation deviation according to shaakha, mss. interpretation, and other matters. I am indebted to Michael Witzel for pointing me to how to go about clarifying this, offering several points for further review. For a more detailed discussion with links and references, click here. Subsequent to this, I have preserved the entire accent scheme so carefully encoded at Texas, and a great deal of information has been preserved behind the otherwise enigmatic ampersand markings.
It also becomes apparent, per Gonda (Mantra Interpretation in the Shatapatha BraahmaNa ) how the ShB represents a later redaction of an earlier text to which both this, the maadhyandina, and the kaaNviya editions refer. The TITUS project, in Frankfurt, Germany, headed by Dr. Jost Gippert, is working on a kaaNviya edition.
Eggeling's ZB for the Sacred Books of East series was often the most reliable, though frequently there were citations missing which, for instance, Griffith's edition identified. Typo's, however, were a problem, for instance viz. 126.96.36.199, which quotes RV 9.66.19c, was marked as 9.16.19 (9.16 has only 8 shloka's). However, Eggeling's citations were rarely wrong while Griffith's frequently were.