1:
2:
3: % ****** Start of file apssamp.tex ******
4: %
5: % This file is part of the APS files in the REVTeX 4 distribution.
6: % Version 4.0 of REVTeX, August 2001
7: %
8: % Copyright (c) 2001 The American Physical Society.
9: %
10: % See the REVTeX 4 README file for restrictions and more information.
11: %
12: % TeX'ing this file requires that you have AMS-LaTeX 2.0 installed
13: % as well as the rest of the prerequisites for REVTeX 4.0
14: %
15: % See the REVTeX 4 README file
16: % It also requires running BibTeX. The commands are as follows:
17: %
18: % 1) latex apssamp.tex
19: % 2) bibtex apssamp
20: % 3) latex apssamp.tex
21: % 4) latex apssamp.tex
22: %
23: %\documentclass[twocolumn,showpacs,preprintnumbers,amsmath,amssymb,prl,endnote]{revtex4}
24: \documentclass[preprint,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
25:
26: % Some other (several out of many) possibilities
27: %\documentclass[preprint,aps]{revtex4}
28: %\documentclass[preprint,aps,draft]{revtex4}
29: %\documentclass[prb]{revtex4}% Physical Review B
30:
31: \def\beq{\begin{equation}}
32: \def\eeq{\end{equation}}
33: \newcommand{\elkobar}[1]{\overset{{}^{{}^{\boldsymbol{\neg}}}}{\smash[t]{#1}}}
34: \usepackage{graphicx}% Include figure files
35: \usepackage{dcolumn}% Align table columns on decimal point
36: \usepackage{bm}% bold math
37:
38: %\nofiles
39:
40: \begin{document}
41:
42: %\preprint{APS/123-QED}
43:
44: \title{MiniBooNE and a $\mathbf{(CP)^2 = -\openone} $ sterile neutrino}
45: % Force line breaks with \\
46:
47: \author{D. V. Ahluwalia-Khalilova}
48: \email{dharamvir.ahluwalia-khalilova@canterbury.ac.nz}
49: \author{Alex B. Nielsen}
50: \email{abn16@student.canterbury.ac.nz}
51: \affiliation{%
52: Department of Physics and Astronomy, Rutherford Building, University of Canterbury, Private Bag 4800, Christchurch 8020, New Zealand}
53:
54:
55: \date{16 February 2007, Accepted for publication in Mod. Phys. Lett. A
56: }% It is always \today, today,
57: % but any date may be explicitly specified
58:
59:
60: \begin{abstract}
61: It has been taken as granted that the
62: observation of two independent mass-squared
63: differences necessarily fixes the number of underlying mass eigenstates
64: as three, and that the addition of a sterile neutrino provides an additional
65: mass-squared difference. The purpose of this Letter is to argue that
66: if one considers
67: a sterile neutrino component that belongs to the $(CP)^2 = -\openone$
68: sector, then both of the stated claims are false. We also
69: outline how the results reported here, when combined with
70: the forthcoming MiniBooNE data and other
71: experiments, can help settle the issue of the $CP$ properties of the sterile
72: neutrino; if such a component does indeed exist.
73: \end{abstract}
74:
75: \pacs{14.60.Pq, 11.30.Er}% PACS, the Physics and Astronomy
76: % Classification Scheme.
77: %\keywords{Suggested keywords}%Use showkeys class option if keyword
78: %display desired
79: \maketitle
80:
81: To understand the excess $\overline{\nu}_e$ events at
82: LSND~\cite{Athanassopoulos:1996jb}, and their absence at
83: KARMEN~\cite{Armbruster:1998uk}, is one of the outstanding issues of
84: neutrino-oscillation phenomenology. On the experimental side, the
85: forthcoming results from MiniBooNE~\cite{Stancu:2006gv,Ray:2007yd}
86: should shed significant light on this issue. On the theoretical side,
87: it has been generally accepted that the observation of two independent
88: mass-squared differences necessarily fixes the number of underlying
89: mass eigenstates to three, and that the addition of a sterile neutrino
90: provides an extra mass-squared difference. The purpose of this Letter
91: is to argue that if one considers a sterile neutrino component
92: belonging to the $(CP)^2 = -\openone$ Wigner sector~\cite{Wigner1962},
93: then both of the stated claims are false. We also outline how the
94: existing data, and the forthcoming results from MiniBooNE
95: and other experiments, can help
96: settle the issue of the $CP$ properties of the sterile neutrino; if
97: such a component does indeed exist.
98:
99:
100: That $(CP)^2= -\openone$ classes exist for spin one half, and that
101: these remain fermionic, is a classic result obtained by Wigner in the
102: early sixties~\cite{Wigner1962}, and later by
103: Lounesto~\cite{Lounesto1997}. This happens without going beyond the
104: Poincar\'e symmetries. Recently, one such quantum field has been
105: explicitly
106: constructed~\cite{Ahluwalia-Khalilova:2004sz,Ahluwalia-Khalilova:2004ab},
107: and its various properties have been investigated by a number of
108: authors~\cite{daRocha:2005ti,Boehmer:2006qq,Boehmer:2007dh}. For
109: instance, the explicit construct has been shown to carry limited
110: interactions with standard model fields. In the absence of a
111: preferred direction \textemdash~such as the one which may arise due to
112: an external magnetic field \textemdash~it carries a Klein-Gordon
113: propagator. Here, however, we do not restrict ourselves to the
114: $(CP)^2 = -\openone$ states reported in
115: ~\cite{Ahluwalia-Khalilova:2004sz,Ahluwalia-Khalilova:2004ab} but
116: allow for a more general identification. The basic inspiration to take
117: from all this is that Dirac, or Majorana, eigenstates do not exhaust
118: the possibilities that exist for a spin one half fermionic field, but
119: that additional, equally fundamental, constructs exist and that these
120: carry new and unexpected physical properties. These should be
121: experimentally investigated without theoretical prejudices. Neutrino
122: oscillations, as will become apparent in what follows, provide an
123: ideal laboratory to probe these aspects of physical reality.
124:
125: Towards the stated goal, as an ansatz, we consider a sterile neutrino
126: that belongs to $(CP)^2 = -\openone$ sector of the spin-$\frac{1}{2}$
127: fermionic Wigner classes~\cite{Wigner1962}, and define a set of four
128: flavour neutrino states~\footnote{Such a suggestion has already been
129: noted by Bernardini in Ref.~\cite{Bernardini:2006cn}.}
130: \beq
131: \vert \xi \rangle = \sum_{j=1,2,3}
132: U_{\xi j}\vert d_j, h \rangle + U_{\xi 4}
133: \vert 4, h \rangle \label{eq:fi}
134: \eeq
135: where the flavour $\xi=\alpha,\beta,\gamma,\delta$. The $U_{\xi j}$
136: and $U_{\xi 4}$ are elements of a $4\times 4$ real unitary matrix
137: carrying no CP violation. This assumption is merely for simplicity and
138: may be relaxed \footnote{By adding the $(CP)^2 = \pm \openone$ sectors
139: together, however, we do allow for the possibility of CP
140: violation. This may lead to mixing of the $(CPT)^2 = \pm \openone$
141: Wigner classes and to a novel form of $CPT$ violation. The latter is
142: presently widely considered in the context of neutrino-oscillation
143: data, see, e.g., Ref.~\cite{Murayama:2000hm,Ahluwalia:1998xb}. For
144: additional discussion of this point, in a related context, the reader
145: is referred to
146: Ref.~\cite{Ahluwalia-Khalilova:2004sz,Ahluwalia-Khalilova:2004ab}}.
147: The $\vert d,h \rangle$ are Dirac mass eigenstates, with $(CP)^2 = +
148: \openone$; while $\vert 4,h \rangle$ is a sterile mass eigenstate,
149: with $(CP)^2 = - \openone$. Each of the four mass eigenstates carries
150: a different mass, and $h$ denotes helicity. The assumption that
151: $(CP)^2=-\openone$ for the sterile component guarantees that it is not
152: a Dirac mass eigenstate. Apart from the stated $(CP)^2$ properties,
153: the introduced Dirac and sterile sectors carry the following
154: additional $CP$ properties
155:
156: \vspace{5pt}
157: \noindent
158: $\mathbf{Dirac},\;\{C,P\}=0:$
159: \begin{eqnarray}
160: CP \vert d, h\rangle = \vert\bar{d},-h\rangle,\;
161: CP \vert \bar{d}, h\rangle = \vert d,-h\rangle.
162: \end{eqnarray}
163: $ \mathbf{Sterile},\;[C,P]=0:$
164: \begin{eqnarray}
165: && CP \vert 4, h \rangle = - i \vert \elkobar{4}, -h\rangle,\;
166: CP \vert 4, - h \rangle = + i \vert \elkobar{4}, h\rangle \\
167: && CP \vert \elkobar{4}, h \rangle = - i \vert 4, -h\rangle,\;
168: CP \vert \elkobar{4}, - h \rangle = + i \vert 4, h\rangle.
169: \end{eqnarray}
170:
171: The former are in accord with the careful treatment given in
172: Ref.~\cite{Weinberg1995}. The latter, after appropriate
173: re-identifications, coincide with the $(CP)^2=-\openone$ construct of
174: Ref.~\cite{Ahluwalia-Khalilova:2004sz,Ahluwalia-Khalilova:2004ab};
175: where, incidentally, it corresponds to the $(CPT)^2=-\openone $ Wigner
176: class. For the chosen sterile sector, $CP$ is an anti-unitary
177: operator. The $\vert\bar d\rangle$ represents the CP conjugate of a
178: Dirac mass eigenstate $\vert d\rangle$; while generically
179: $\vert\elkobar{\lambda}\rangle $, with $\lambda=4$ above, denotes the CP
180: conjugate of a mass eigenstate with $(CP)^2 = -\openone$.
181:
182: Using the above enumerated results, the successive action of $CP$ on
183: the set of flavour eigenstates (\ref{eq:fi}) yields the following flavour
184: states
185:
186: \begin{eqnarray}
187: &&\hspace{-21pt} \vert \tilde{\xi} \rangle := CP \vert \xi \rangle
188: =
189: \sum_{j} U_{\xi j}\vert \bar{d}_j, - h \rangle - i U_{\xi 4}
190: \vert \elkobar{4}, - h \rangle \label{eq:fi2}\\
191: &&\hspace{-21pt} \vert \xi^\prime \rangle := CP \vert \tilde{\xi} \rangle
192: = \sum_{j}
193: U_{\xi j}\vert d_j, h \rangle - U_{\xi 4}\vert 4,
194: h \rangle \label{eq:fi3} \\
195: && \hspace{-21pt}
196: \vert \tilde{\xi}^\prime \rangle := CP \vert \xi^\prime
197: \rangle
198: =
199: \sum_{j} U_{\xi j}\vert \bar{d}_j, - h \rangle + i U_{\xi 4}
200: \vert \elkobar{4}, - h \rangle \label{eq:fi4}
201: \end{eqnarray}
202: with $CP \vert \tilde{\xi}^\prime \rangle$ being identical to the
203: original flavour $\vert \xi \rangle$ defined in Eq.~(\ref{eq:fi}).
204: Here, and in the following,
205: $
206: j,k:=
207: 1,2,3
208: $
209: corresponds to the three Dirac mass eigenstates.
210: The flavour index $\xi$, as already mentioned, runs over four flavours
211: \beq
212: \xi:= \alpha,\beta,\gamma,\delta \label{eq:xiz}.
213: \eeq
214: These flavours would soon be seen to
215: be connected to, but not identical with, the
216: operationally defined flavours of neutrinos.
217: The circumstance that
218:
219: \beq
220: \vert \xi \rangle \stackrel{CP}{\rightarrow}
221: \vert \tilde{\xi} \rangle \stackrel{CP}{\rightarrow}
222: \vert \xi^\prime \rangle \stackrel{CP}{\rightarrow}
223: \vert \tilde{\xi}^\prime \rangle \stackrel{CP}{\rightarrow} \vert \xi \rangle
224: \label{eq:4cp}
225: \eeq
226: forces upon us the fundamental question:
227: \begin{quote}
228: How are the above-obtained flavour states related to the flavour
229: eigenstates $\vert \nu_\ell\rangle$ and $\vert\bar\nu_\ell\rangle$,
230: where $\ell$ corresponds to the operationally-defined flavours
231: $\ell:=e,\mu,\tau,\zeta$
232: \footnote{The new flavour $\zeta$
233: is just a logical necessity of the $4\times 4$ framework. It
234: complements the three flavours associated with electron, muon, and tau
235: charged leptons.}?
236:
237: \end{quote}
238: To answer this question we first re-write (\ref{eq:4cp}) by collecting
239: together flavours connected by $(CP)^2$ \footnote{In the absence of
240: the considered sterile component the flavours connected by $(CP)^2$
241: are identical. Now this is no longer so.}. This gives rise to a pair
242: of flavour sets
243: \beq
244: \left\{
245: \begin{array}{ll}
246: \vert \xi \rangle \\
247: \vert \xi^\prime \rangle
248: \end{array}
249: \right\},\quad
250: \left\{
251: \begin{array}{ll}
252: \vert \tilde\xi \rangle \\
253: \vert \tilde\xi^\prime \rangle
254: \end{array}
255: \right\}\label{eq:fset}.
256: \eeq
257: The action of $CP$ is now encoded in the following schematic equation
258: \beq
259: CP \left\{
260: \begin{array}{ll}
261: \vert \xi \rangle\\
262: \vert \xi^\prime \rangle
263: \end{array}
264: \right\}\rightarrow
265: \left\{
266: \begin{array}{ll}
267: \vert \tilde\xi \rangle\\
268: \vert \tilde\xi^\prime \rangle
269: \end{array}
270: \right\},\;\;
271: CP \left\{
272: \begin{array}{ll}
273: \vert \tilde\xi \rangle\\
274: \vert \tilde\xi^\prime \rangle
275: \end{array}
276: \right\}\rightarrow
277: \left\{
278: \begin{array}{ll}
279: \vert \xi^\prime\rangle\\
280: \vert \xi \rangle
281: \end{array}
282: \right\}.
283: \eeq
284: The action of $CP$ thus rotates between the two flavour sets.
285: This circumstance leads us to suggest the following
286: identification
287:
288: \begin{quote}
289: In any given production of $\nu_\ell$ one either creates a
290: $\vert\xi\rangle$ or $\vert\xi^\prime\rangle$ with equal
291: probability. Similarly, $\bar\nu_\ell$ creation corresponds to the
292: production of either $\vert\tilde\xi\rangle$ or
293: $\vert\tilde\xi^\prime\rangle$ with equal probability.
294:
295: \end{quote}
296: So, for example,
297: \beq
298: \bar\nu_\mu:= \left\{
299: \begin{array}{ll}
300: \vert \tilde\beta \rangle \\
301: \vert \tilde\beta^\prime \rangle
302: \end{array}
303: \right\}, \quad
304: \bar\nu_e:= \left\{
305: \begin{array}{ll}
306: \vert \tilde\alpha \rangle \\
307: \vert \tilde\alpha^\prime \rangle\label{eq:fdef}
308: \end{array}
309: \right\}.
310: \eeq
311: The discussion above constitutes our working answer to the asked
312: question. At this stage it is an hypothesis. Its validity, or its
313: final acceptance, should be deferred to experiments, and/or to
314: additional theoretical work. What appears certain is that if nature
315: does superimpose mass eigenstates with differing $(CP)^2$, or for that
316: matter with different $(CPT)^2$ properties, then the
317: particle-antiparticle concept must undergo a fundamental
318: reexamination. Our suggestion constitutes a preliminary, and perhaps
319: first, theoretical attempt in that direction. However, its flavour
320: mirrors the remarks contained in the opening
321: paragraph of Sec. III A of Langacker-London paper on nonorthogonal
322: neutrinos~\cite{Langacker:1988up}.\footnote{The idea of non-orthogonal neutrinos
323: seems to have first arisen in Ref.~\cite{Lee:1977qz}.}
324:
325: One of our tasks here is to understand
326: excess $\overline{\nu}_e$ events at
327: LSND~\cite{Athanassopoulos:1996jb}, their absence at
328: KARMEN~\cite{Armbruster:1998uk}, and the forthcoming results from the
329: MiniBooNE~\cite{Stancu:2006gv,Ray:2007yd}.
330: As such, we now exploit the emergent interpretation for calculating
331: the flavour oscillation probability for $\bar\nu_\mu \to \bar\nu_e$.
332: It is given by
333: \begin{eqnarray}
334: \hspace{-21pt}\mathcal{P}\left(\bar\nu_\mu\to\bar\nu_e\right)&=&
335: \frac{1}{4}\bigg\{
336: \Big\vert\left\langle\tilde\alpha\Big\vert\tilde\beta\right
337: \rangle^t\Big\vert^2
338: +\Big\vert\left\langle\tilde\alpha^\prime\Big\vert\tilde\beta
339: \right\rangle^t\Big\vert^2 \nonumber\\
340: &&\hspace{11pt} +\Big\vert\left\langle\tilde
341: \alpha\Big\vert\tilde\beta^\prime\right\rangle^t\Big\vert^2
342: +\Big\vert\left\langle\tilde\alpha^\prime\Big\vert\tilde\beta^\prime
343: \right\rangle^t\Big\vert^2
344: \bigg\}\label{eq:pmbeb}
345: \end{eqnarray}
346: where the notation $\vert ~\rangle^t$ corresponds to the space-time evolved
347: state in the usual neutrino-oscillation experimental setting.
348: In order to understand the ensuing expression for the
349: $\mathcal{P}\left(\bar\nu_\mu\to\bar\nu_e\right)$, we would now like
350: to present each of the four terms explicitly. To facilitate this, we
351: introduce the following definitions
352:
353: \begin{eqnarray}
354: &&\mathcal{A}_{jk} := 4 U_{\alpha j} U_{\alpha k} U_{\beta j} U_{\beta k}
355: \label{eq:ajk} \\
356: && \mathcal{\chi} := \frac{1}{2}\bigg\{\sum_{j}
357: U_{\alpha j} U_{\beta j} - U_{\alpha 4} U_{\beta 4} \bigg\}^2
358: \label{eq:apm}\\
359: && \varphi_{jk}:=\frac{\Delta m_{jk}^2 L}{4 \hbar c E} = \frac{1.27
360: \Delta m_{jk}^2
361: (\mathrm{eV^2}) L (\mathrm{km})}{E(\mathrm{GeV})} \label{eq:dmjk}
362: \end{eqnarray}
363: where $L$ refers to the source-detector distance, $E$ is the neutrino energy,
364: and $\Delta m_{jk}^2:= m_j^2-m_i^2$.
365: We also need to define $\mathcal{A}_{j4}$ and
366: $\Delta m_{j4}^2$. These are obtained by the replacement
367: $k \to 4$ in Eq. (\ref{eq:ajk}) and Eq. (\ref{eq:dmjk}),
368: respectively.
369:
370:
371: With these definitions, the four terms
372: that appear in Eq.~(\ref{eq:pmbeb}) read
373: \begin{eqnarray}
374: && \hspace{-21pt} \Big\vert\left\langle\tilde\alpha\Big\vert\tilde\beta\right
375: \rangle^t\Big\vert^2 = \Big\vert\left\langle\tilde\alpha^\prime
376: \Big\vert\tilde\beta^\prime\right\rangle^t\Big\vert^2
377: \nonumber \\
378: && =
379: - \sum_{j<k}\mathcal{A}_{jk}\sin^2 \varphi_{jk}
380: -\sum_{j} A_{j 4} \sin^2 \varphi_{j4} \nonumber
381: \\
382: &&\hspace{-21pt} \Big\vert\left\langle\tilde\alpha^\prime\Big\vert\tilde\beta
383: \right\rangle^t\Big\vert^2 =
384: \hspace{11pt} \Big\vert\left\langle\tilde
385: \alpha\Big\vert\tilde\beta^\prime\right\rangle^t\Big\vert^2
386: \nonumber \\
387: && =
388: 2 \chi
389: - \sum_{j<k}\mathcal{A}_{jk}\sin^2 \varphi_{jk}
390: +\sum_{j} A_{j 4} \sin^2\varphi_{j4}.\nonumber
391: \end{eqnarray}
392: As such, we have the central result of the Letter
393:
394: \beq
395: \mathcal{P}\left(\bar\nu_\mu\to\bar\nu_e\right)
396: = \chi
397: - \sum_{j< k}\mathcal{A}_{jk}\sin^2 \varphi_{jk}.
398: \eeq
399: It is to be immediately noted that due to the manifest cancellation of
400: the $\varphi_{j4}$ terms, the $(CP)^2 = -\openone$ sterile-neutrino
401: component does \textit{not} induce any oscillatory terms. Its
402: presence is felt solely through the constant \textit{flavour
403: transmutation} term. The transmutation term may be re-written as
404: \beq
405: \chi =
406: \bigg\{ \sum_{j}
407: U_{\alpha j} U_{\beta j} \bigg\}^2 + \big\{U_{\alpha 4} U_{\beta 4}\big\}^2
408: \label{eq:ft}.
409: \eeq
410: It represents a fundamentally unavoidable $\bar\nu_e$
411: ``contamination'' in a $\bar\nu_\mu$ beam.
412: We refer to this
413: ``contamination'' as a flavour transmutation rather than a
414: flavour oscillation due to its $L/E$ independence.
415: Its origin can be traced back to the fact that
416: while each of the flavours,
417: $\alpha,\beta,\gamma,\delta$ in (\ref{eq:fi2})
418: are mutually orthogonal (e.g., $\langle\tilde\alpha\vert\tilde\beta\rangle = 0$),
419: and the same being true with the flavours, say, in
420: (\ref{eq:fi4}) (e.g., $\langle\tilde\alpha^\prime\vert\tilde\beta^\prime\rangle = 0$);
421: the
422: flavour states
423: such as $\vert \tilde\alpha\rangle$ and
424: $\vert \tilde\beta^\prime \rangle$
425: have non-vanishing overlap, i.e. $\langle\tilde\alpha\vert\tilde \beta^\prime
426: \rangle \ne 0$.
427:
428: The above remarks parallel those surrounding Eq.~(25) of
429: the Langacker-London paper on nonorthogonal
430: neutrinos~\cite{Langacker:1988up}. There, the effect
431: arises from a mismatch between the light and heavy neutrinos.
432: In our case a very similar result arises from a
433: mismatch between the $(CP)^2$ properties
434: of the Dirac and sterile mass eigenstates.
435: We do not assume that the mass of the sterile component is
436: significantly larger than the mass of the Dirac components.
437:
438:
439: Following the same procedure as above, we have also calculated
440: $\mathcal{P}\left(\bar\nu_\mu\to\bar\nu_\mu\right)$,
441: $\mathcal{P}\left(\bar\nu_\mu\to\bar\nu_\tau\right)$,
442: $\mathcal{P}\left(\bar\nu_\mu\to\bar\nu_\zeta\right)$. These, together
443: with $\mathcal{P}\left(\bar\nu_\mu\to\bar\nu_e\right)$, sum to unity,
444: and carry no oscillatory term associated with the $(CP)^2 = -\openone$
445: sterile-neutrino component. Each of these contains a flavour
446: transmutation term. We have also calculated
447: $\mathcal{P}\left(\nu_\mu\to\nu_e\right)$ and find it equals
448: $\mathcal{P}\left(\bar\nu_\mu\to\bar\nu_e\right)$.
449:
450:
451: It is now to be noted that a non-zero $\chi$ mimics an anomalous
452: decay that produces a $\bar\nu_e$. For this reason existing analysis
453: of the KARMEN and LSND data may be used to determine the value of
454: $\chi$. Specifically, KARMEN saw $15$ $\bar\nu_e$ like
455: sequences. These are in agreement with the background expectation of
456: $15.8 \pm 0.5$ $\bar\nu_e$ like
457: sequences~\cite{Reichenbacher:2005nc}. If the LSND result is
458: assumed to be entirely due to an anomalous decay, then KARMEN should
459: have seen $15$ additional $\bar\nu_e$ like
460: events~\cite{Goldman:2007pc}. But because
461: a non-zero $\chi$ mimics an anomalous decay,
462: the absence of the additional events at
463: KARMEN places severe constraints
464: on $\chi$. Therefore,
465: unless MiniBooNE reports
466: a confirmation of a LSND-like signal
467: with essentially
468: no $L/E$ dependence, the existing data on KARMEN
469: rules out a non-zero $\chi$; i.e., it excludes a $(CP)^2 = -\openone$
470: sterile neutrino component
471: in the presented $3+1$ scenario.
472:
473:
474: The central result of this Letter establishes that the addition of
475: a
476: $(CP)^2 = - \openone$ sterile neutrino component does not introduce an
477: extra mass-squared difference in the neutrino oscillation probability.
478: It is, therefore, true that the observation of two independent
479: mass-squared differences does not necessarily fix the number of
480: underlying mass eigenstates as three. A $3+2$ extension of our formalism
481: yields only $3$, not $4$, oscillatory terms in the relevant flavour oscillation
482: probabilities.
483: For these reasons we argue that the
484: $(CP)^2 = \pm \openone$ nature of any sterile neutrino component is
485: experimentally accessible.
486:
487:
488:
489: \begin{acknowledgments}
490: We wish to thank Terry Goldman, Daniel Grumiller, and
491: Ben Leith for helpful discussions and comments.
492: We are grateful to an anonymous referee for bringing to
493: our attention the subject of nonorthogonal neutrinos~\cite{Langacker:1988up}.
494:
495: \end{acknowledgments}
496: \bibliography{an}
497:
498: \end{document}
499: