Case law you can use: Examiner can't pick and choose unrelated embodiments in a §102

What do you do when an Examiner gives you a §102 rejection that combines completely unrelated teachings in a single reference?

With a §103, at least you can argue why these two pieces are unrelated and thus would not be combined. OK, you're unlikely to convince the Examiner with this argument — but at least it's legally relevant. But "wouldn't make sense to combine" isn't legally relevant to a §102...is it?

Not exactly. But there is another weapon you can use:

The [prior art] reference must clearly and unequivocally disclose the claimed [invention] or direct those skilled in the art to the [invention] without any need for picking, choosing, and combining various disclosures not directly related to each other by the teachings of the cited reference.
In re Arkley, 455 F.2d 586, 587 (CCPA 1972).

Although In re Arkley was a chemical compound case, the Board does ocassionally use it to overturn anticipation rejections in non-chemical cases.

On example is Ex parte Ludwig (Appl. No. 09/812,400). The anticipatory reference taught two embodiments of a tone generator: FIG. 2 was a hardware implementation; FIG. 4 was a software implementation. The Examiner read one claim limitation on a component in FIG. 2 and another claim limitation on a component in FIG. 4. The Board reversed, holding that "the Examiner’s anticipation rejection cannot be based on merely picking and choosing multiple, distinct teachings from Suzuki that could somehow be combined to arrive at the claimed invention. See [Net MoneyIn, 545 F.3d ] at 1371. See also In re Arkley, 455 F.2d at 587. Rather, to anticipate, Suzuki must disclose every recited element arranged as in the claim—which it does not." (Decision, pp. 10-11.)

Another example is Ex parte Sato (Appl. No. 10/791,829). The claims were directed to an electro-deionization apparatus which introduced deionized water from desalting compartments into a specific location in concentrating compartments or allowed concentrated water to flow out of a specific location in the concentrating compartments. The Board overturned the rejection, citing In re Arkley and holding that the "[t]he Examiner has combined the Liang ‘037 Figures 1 and 13 without establishing a direct relation of the relied-upon features." (Decision, p. 5.)

Here's another example: Ex parte Omshehe (Appl. No. 09/954,509). The claims were directed to verifying software licenses when executing software. The anticipatory reference taught two distinct licensing mechanisms. The Examiner relied on one of these mechanisms to teach some claim limitations, then on the other mechanism to teach the remainder of the limitations. The Board reversed the rejection, holding that "such a mix[ing] and match[ing] [of] elements from Redding's two distinct authorization modes is inappropriate for a rejection based on anticipation. Therefore, the Examiner fails to show every element of the claimed invention arranged as in the claims." (Decision, p. 6.) Interestingly, the Board did not cite any caselaw for the proposition that mixing and matching is inappropriate. But clearly In re Arkley is appropriate.

As a final example, look at Ex parte Strongin (Appl. No. 09/825,905). The technology involved (computer processor architecture) is more complicated than I want to get into. But the takeaway is that the Board reversed, citing In re Arkley and stating that "The Examiner is apparently of the opinion that the claimed feature of (1) controlling access to selected information using a first table is taught by Figure 45 of Nozue, and (2) controlling access to selected information using a second table is taught by Figure 24A of Nozue. We find that Figures 24A and 45 depict different embodiments of the Nozue reference." (Decision, pp. 6-7.)